JP4788394B2 - Image processing apparatus, image processing method, and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program that adjust the brightness of image data.

  Image data created by an image input device such as a digital still camera is created based on a captured image, and a color image is made up of a plurality of element colors for each pixel, for example, RGB (red, green, blue). The data is expressed in gradation. In the image input apparatus, estimation processing such as estimating a light source based on a captured image is performed, and image quality adjustment such as exposure adjustment is automatically performed as appropriate.

Further, in paragraph 0087 of Patent Document 1, a correction amount Th to be corrected is calculated according to the calculated “preferred color target value” and the characteristic value obtained from the characteristic value table of the original image data. It is described that image data after correction is obtained by performing color correction on the image data by a correction amount. As described in paragraphs 0088-0089 of the same document, the correction amount Th is set such that ΔS _S is an area occupancy of a specific color (skin color), a and r are constants of 0 to 1, and 0 ≦ ΔS _S ≦ a. In this case, a (1-r) × (ΔS _S ) ^r , 1− (1-a) ^1−r × (1−ΔS _S ) ^{r in} the case of a ≦ ΔS _S ≦ 1, and a specific color in the image data The correction amount is set in accordance with the area occupation ratio ΔS of the region (skin color region).
Therefore, for the specific color area (skin color area) of the original image data, the color correction is uniformly applied to the “preferred color target value” by the correction amount Th regardless of the color of the background area.
JP 2004-192614 A

  Examination of various image data created by the image input device revealed that there were cases where appropriate exposure adjustment was not performed by the image input device. Therefore, it has been desired to appropriately adjust the exposure of the image data created by the image input device even when appropriate exposure adjustment is not performed by the image input device.

  An object of the present invention is to appropriately adjust the brightness of image data created by an image input device.

Means for Solving the Problems and Effects of the Invention

  In order to achieve the above object, the present invention provides an area extracting means for extracting a predetermined color area from image data created by an image input device, and an area brightness for acquiring the brightness of the color area from the extracted color area. An image processing apparatus comprising: an acquisition unit; and an adjustment unit configured to perform brightness correction on the image data so that the brightness of the color area is set to a target brightness set in advance for the color area. Brightness adjustment means, wherein the adjustment means is such that the brightness of the color region is darker than the first brightness darker than the target brightness and the background brightness is lower than the second brightness darker than the target brightness. When the image is dark, brightness correction is performed on the image data with brightness lower than the target.

  In the background brightness acquisition means, a representative color of the background of the color region is extracted from the image data. Then, when the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the target brightness is lowered and the brightness correction of the image data is performed by the adjusting means. Done.

Examining various image data created by the image input device, when the appropriate brightness adjustment is not performed by the image input device, a very dark image is created due to insufficient strobe light, etc. There was a case. If brightness correction is performed on an image to uniformly set the brightness of the color area to the target brightness, noise will appear in the image of an extremely dark image as a whole, and the image quality after image processing will not improve as intended. Sometimes. Even if the specific color area (skin color area) is uniformly color-corrected to the “desired color target value” by the correction amount Th regardless of the color of the background area as in the technique described in Patent Document 1, it is very much overall. Noise appears in the dark image. In addition, when an image is captured in a very dark environment, a person tends to feel less unnaturalness with respect to darkness. Therefore, for a very dark image as a whole, the image quality can be improved by lowering the brightness correction target in the dark direction.
In the present invention, when the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the target brightness is lowered, so that the above-described problems can be solved. . Therefore, according to the present invention, it is possible to appropriately adjust the brightness of the image data created by the image input device even when the appropriate brightness adjustment is not performed by the image input device.

The image data can be, for example, image data expressing a color image with a color component amount for each pixel. The number of pixels may be any number as long as it can represent an image, and may have a configuration of a plurality of pixels. For example, a small image such as 8 × 8 pixels may be represented. The color component amount can be a value representing the number of each color component in a color space having a plurality of color components, for example.
The predetermined color area may be a face area, a skin color area, a sky blue area, or the like, and may be an area impressed by many people, an area related to color reproduction that humans feel preferable, or the like. The target brightness can be the brightness that humans feel desirable, the brightness of the memory color, the brightness that does not depend on human senses, etc., and the brightness that changes according to the brightness of the color area or background However, the brightness may be constant regardless of the brightness of the color area or the background. The background of the color area may be all areas except the color area from the entire image, or may be a partial area of the area. The brightness is represented by the brightness, brightness, average of color component values, and the like of a color space composed of a color coordinate plane and a brightness axis. The brightness of the color area or background is represented by a simple average or a weighted average of values representing the brightness of each pixel in the color area or background. The brightness correction may be performed uniformly on the entire image data, or only the extracted color region may be performed.
The color space composed of the color coordinate plane and the brightness axis includes a CIE L ^* a ^* b ^* color space and a CIE L ^* u ^* v ^* color space defined by the International Lighting Commission (CIE). Here, L ^* is a color component representing lightness (brightness), and a ^* , b ^* , u ^* , v ^* are color components representing hue and saturation. Hereinafter, “ ^* ” is omitted.
Note that decreasing the brightness means decreasing the brightness, for example, by decreasing the brightness.

The target brightness may be a brightness that changes according to a difference between the brightness of the background and the brightness of the color region. Since the sensory brightness of the color area perceived by a person changes depending on the difference from the brightness of the background, the brightness of the image created by the image input device can be adjusted more appropriately.
More suitably, when the target brightness is determined to compensate for a difference in sensory brightness of the color area perceived by a person according to a difference between the brightness of the background and the brightness of the color area It becomes possible to adjust the brightness of the image.
Here, the difference is, for example, the difference between the value representing the brightness of the background and the value representing the brightness of the color region, the ratio of the value representing the brightness of the background and the value representing the brightness of the color region, and 1 And the difference between

Further, when the lightness of the color region is lower than the first lightness lower than the target lightness and the lightness of the background is lower than the second lightness lower than the target lightness, the lightness of the background The target brightness that increases continuously or stepwise as the brightness difference ΔL obtained by subtracting the brightness of the color area increases within a predetermined range may be lowered within a range higher than the brightness of the color area. This also makes it possible to adjust the brightness of the image more appropriately.
When the lightness takes discrete values, the target lightness may be increased stepwise so that all of the discrete values can be obtained as the lightness difference ΔL increases within a predetermined range. Alternatively, it may be increased stepwise at larger intervals.

  By the way, when the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the brightness of the color area is within a range brighter than the brightness of the color area. The brightness of the target may be lowered so that the brightness increases continuously or stepwise as the brightness increases. Further, when the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the target brightness is within a range brighter than the brightness of the color area. And the target brightness when the brightness of the color area is brighter than the third brightness that is darker than the first brightness, the target brightness when the brightness of the color area is darker than the third brightness. The brightness may be higher than the target brightness. Then, it becomes possible to brighten the dark image as a whole so that noise does not appear and appropriately improve the image quality.

In addition, the present invention is an image processing apparatus that includes the region extraction unit, the region brightness acquisition unit, and an adjustment unit. The image processing apparatus includes the background brightness acquisition unit, and the adjustment unit determines whether the background brightness is high. When the brightness of the color area is brighter and the difference between the brightness of the background and the brightness of the color area is larger than a predetermined difference, brightness correction is performed on the image data with brightness lower than the target. It is characterized by.
That is, when the brightness of the background is brighter than the brightness of the color area and the difference between the brightness of the background and the brightness of the color area is larger than the predetermined difference, the target brightness is lowered and the brightness of the image data is increased. Correction is performed.
When various image data created by the image input device was examined, there was a case where an image was created by taking an image under a backlight condition as a case where appropriate brightness adjustment was not performed by the image input device. When brightness correction is performed on an image to uniformly set the brightness of the color area to the target brightness, noise appears in the image created under backlight conditions, and the image quality after image processing is improved as intended There are things that do not. As in the technique described in Patent Document 1, the specific color area (skin color area) can be uniformly corrected to the “desired color target value” by the correction amount Th regardless of the color of the background area. In the created image, noise appears in the image. Therefore, in an image created under backlight conditions, the image quality can be improved by lowering the brightness correction target in the dark direction.
In the present invention, when the brightness of the background is brighter than the brightness of the color area and the difference between the brightness of the background and the brightness of the color area is larger than a predetermined difference, the target brightness is lowered, so the above-described problems are solved. be able to. Therefore, according to the present invention, it is possible to appropriately adjust the brightness of the image data created by the image input device even when the appropriate brightness adjustment is not performed by the image input device.

Here, when the brightness of the background is brighter than the brightness of the color area, and the difference between the brightness of the background and the brightness of the color area is larger than a predetermined difference, the brightness within the range brighter than the brightness of the color area. The target brightness may be decreased continuously or stepwise as the difference between the brightness of the background and the brightness of the color area increases. Then, the image created under the backlight condition can be brightened to such an extent that noise does not appear, and the image quality can be appropriately improved.
Further, when the lightness difference ΔL obtained by subtracting the lightness of the color area from the lightness of the background is larger than a predetermined lightness difference ΔLth, the lightness difference ΔL increases continuously within a range higher than the lightness of the color area. Alternatively, the lightness of the target may be lowered. This also makes it possible to appropriately improve the image quality by making the image created under backlight conditions bright enough to prevent noise from appearing.

Further, according to the ninth aspect of the present invention, the brightness of the image data can be adjusted more appropriately .

The above-described image processing apparatus includes various modes such as being implemented together with other methods in a state of being incorporated in a certain device. For example, the present invention can be applied to a print control apparatus that performs control for printing image data after image processing, a printing system that further includes a printing apparatus, and the like.
Further, the present invention can also be applied to a control method having a process corresponding to the configuration of the above-described apparatus and a program that causes a computer to realize a function corresponding to the configuration of the above-described apparatus. Inventions such as an image processing method and an image processing program are also possible. Has the same action and effect. Furthermore, the present invention can also be applied as a computer-readable recording medium that records the program. Of course, configurations described in claims 2 to 9 are also applicable to the method, the system, the program, and the recording medium.

Hereinafter, embodiments of the present invention will be described in the following order.
(1) Configuration of printing system including image processing apparatus:
(2) Image processing operations and effects:
(3) Modification:

(1) Configuration of printing system including image processing apparatus:
FIG. 1 is a diagram schematically illustrating image processing according to an embodiment of the present invention. FIG. 2 is a block diagram schematically illustrating the configuration of a print control apparatus U0 including an image processing apparatus according to an embodiment of the present invention. 3 is a block diagram showing an outline of the configuration of a printing system including the printing control apparatus U0. FIG. 4 is a diagram schematically showing a color range R11 as a color area on the ab plane. FIG. FIG. 6 is a diagram schematically illustrating how the region R1 is extracted, FIG. 6 is a diagram schematically illustrating how the sensory brightness J that changes according to the brightness of the background is compensated, and FIG. 7 is the brightness of the color region in a dark image. FIG. 8 and FIG. 9 are flowcharts showing print control processing performed by the print control apparatus U0. FIG. 10 is a diagram schematically showing how the brightness of image data is corrected. 11 is a color adjustment process performed by the digital still camera; Diagram schematically showing a conversion characteristic of FIG. 12 is a block diagram schematically showing a configuration of a digital still camera. The printing system includes a personal computer (PC) 10 which is an image processing apparatus of the present invention, an inkjet printer (printing apparatus) 20 capable of color printing, and the like. Of course, the computer used in the present invention is not limited to a PC.

In the PC 10, the CPU 11 controls the entire PC via the system bus 10a. The bus 10a is connected to semiconductor memories 12, 13, various drives 15, 16, various interfaces (I / F) 17a to 17e, etc., and a hard disk (magnetic disk) 14 is also connected via a hard disk drive. . The image processing program 14a of the present invention is stored in the hard disk (HD) 14, transferred to the RAM 13 by the CPU 11, and executed. The HD 14 also includes information for defining a predetermined color space such as a Lab color space (color space information 14b), information for extracting a color area such as a face area (color extraction information 14c), a color area The target brightness information 14d representing the target brightness, threshold values 14e and f for determining whether or not the image is a dark image such as a strobe failure condition, and whether or not the image is in a backlight condition For a predetermined information storage area (information recording area) for storing (recording) threshold value 14g, brightness information 14h for a corrected target with reduced brightness, color conversion LUT (color conversion table), various threshold values, and the like. Yes. The RAM 13 temporarily stores image data 13a, color area information DA3 indicating the range of the color area on the image data, area brightness information DA4 indicating the brightness of the color area, and the brightness of the background of the color area. Is stored.
This image processing program may be configured by any of an operating system (OS) such as a printer driver, an application program (APL), and OS and APL. In addition to the HD, the medium storing this program may be a CD-ROM 15a, a flexible disk, a semiconductor memory, or the like. Further, the communication I / F 17d may be connected to the Internet network, and the program of the present invention may be downloaded from a predetermined server and executed.

  An image input device such as a digital still camera (DSC) 50 or a digital video camera can be connected to the peripheral device I / F 17a (for example, USB I / F). The CRT I / F 17b is connected with a display 18a for displaying an image corresponding to the data based on the color image data, and the input I / F 17c is connected with a keyboard 18b and a pointing device (mouse 18c) as operation input devices. The printer 20 is connected to the printer I / F 17e via, for example, a serial I / F cable.

  For example, as shown in FIG. 12, the DSC 50 includes an optical unit 52 and a control unit 54. In the illustrated optical unit 52, a controller 52e is connected to a CPU 54c, and an autofocus controller 52d, a CCD unit 52f, an auto gain controller (AGC) 52h, and a strobe 52g are connected to the controller 54, and an autofocus mechanism is connected to the autofocus controller 52d. 52b and a distance measuring unit 52c are connected, a lens 52a is connected to the autofocus mechanism 52b, an A / D conversion circuit 52i is connected to the bus 54a, an AGC 52h is connected to the A / D conversion circuit, and the AGC is connected to the AGC. A CCD unit 52f is connected. In the control unit 54, the CPU 54b, ROM 54c, RAM 54d, LCD panel 54f, operation panel 54e, I / O circuits 54g, h are connected to the bus 54a, and a memory card 54i is connected to the I / O circuit 54g. The PC I / F 17a is connected to the O circuit 54h.

  When the subject is imaged, light from the subject enters the CCD unit 52f via the lens 52a, and an analog electrical signal generated by each imaging element of the CCD unit is sent to the A / D conversion circuit 52h via the AGC 52h. Entered. Then, by the processing of the CPU 54b, a digital value for each image sensor is read from the A / D conversion circuit 52h, and image data in which the captured image is expressed in gradation with a plurality of element colors (for example, RGB) for each pixel is generated. It is stored in the work area of the RAM 54d. The gradation value of each element color of the generated image data is a value that is substantially proportional to the amount of light incident on each image sensor of the CCD unit. The amount of light is defined as the product of the brightness of light incident per unit area of the light receiving surface of each image sensor and the exposure time. If the gradation value is R, the brightness of the R component of the light and the exposure time Product. Therefore, the gradation value of each element color is a value that is substantially proportional to the product of the luminance and the exposure time for each component color component of light.

Here, since the intensity of light perceived by the human eye is not proportional to the amount of light, the DSC 50 converts the gradation value of each element color so as to be proportional to the intensity of light perceived by the human eye. . In addition, the DSC appropriately performs a process for adjusting the exposure and the like based on the image data of the captured image.
The image data subjected to the above-described image quality adjustment processing is stored in the image area of the RAM 54d, and is appropriately output from the I / O circuit 54h to the PC 10 or stored in the memory card 54i.

The lower graph in FIG. 11 shows DSC color adjustment processing characteristics, that is, input / output characteristics corresponding to DSC conversion characteristics. The upper graph in the figure shows input / output characteristics obtained by dividing the characteristics of the color adjustment processing into four. In each graph, the horizontal axis represents an input value and the vertical axis represents an output value, which indicates that each input value is converted into a corresponding output value on the graph according to the characteristics indicated by the solid line. A straight line indicated by a broken line indicates a characteristic when the input value is equal to the output value. In the graph 5, the horizontal axis represents the amount of light and the vertical axis represents the image data 15a. In this embodiment, the input / output characteristics are the same for each of the RGB color components. Therefore, in DSC, it can be considered that the amount of light for each RGB acquired by each CCD image sensor is converted by the input / output characteristics shown in graph 5, and the gradation value for each RGB in the image data is acquired. Here, the amount of light for each RGB is expressed as Y, and the output value based on the conversion characteristics of DSC is expressed as F ⁻¹ (Y). This Y corresponds to the luminance if the exposure time is constant.

  Graphs 1 to 4 are input / output characteristics obtained by dividing the above conversion characteristics in order to extract DSC-specific color adjustment processing. Graphs 1, 3, and 4 convert the color system of gradation values. Meaning adjustment processing, graph 2 is adjustment processing for improving the appearance of the image. The DSC manufacturer performs its own color adjustment process to improve the appearance of the image, but the appearance of the image takes into account how each RGB color is perceived by the human eye. While determined. Therefore, it is assumed that the process of converting the signal according to the amount of light into the signal of the sRGB standard includes a process of improving the appearance with a signal in a state where it is easy to evaluate the perception by the human eye, and the graph The color adjustment process was disassembled as in 1-4. That is, the graph 1 is a characteristic for converting a value proportional to the amount of light into a value proportional to the brightness perceived by the human eye. Graph 2 is a characteristic for converting a value proportional to the brightness perceived by the human eye into a value that improves the appearance of the image. Graphs 3 and 4 are characteristics for converting values that improve the appearance of the image into signals of the sRGB standard.

Specifically, the horizontal axis of the graph 1 is the amount of light Y, and the vertical axis is a value (sL ^* described later) corresponding to the lightness L ^* . That is, when the value of the lightness L ^* increases, the brightness perceived by humans changes in proportion to the change in the value. Therefore, the CIE Lab formula L ^* = 116Y for converting the brightness of the tristimulus values into the lightness L ^{* -1/3} -16 (1)
Can be used to describe the characteristic of converting the amount of light Y into a value corresponding to the lightness L ^* . The DSC acquires a signal indicating the amount of light Y for each RGB color, but if the light amount Y of each RGB color is substituted into Equation (1), a gradation value proportional to the brightness perceived by human beings is converted to RGB. It can be acquired for each color. In the present embodiment, this value is referred to as sL ^* .

The horizontal axis of the graph 2 is the sL ^* , and the vertical axis is a value that has been converted to improve the appearance of the image. In the present embodiment, this value is called pL ^* . This input / output characteristic is determined independently by the DSC manufacturer, but according to the analysis of the applicant of the present application, although there is a slight difference for each manufacturer, the input / output characteristics of each manufacturer are described with substantially the same characteristics. It has been found that it can be done. That is, as shown in the graph 2, an S-curve (pL ^* = G (sL ^* )) that converts an input value to a smaller output value below the halftone and converts the input value to a larger output value above the halftone. ) Can describe input / output characteristics. With this input / output characteristic G (sL ^* ), it was possible to describe a color adjustment process that should be uniquely determined by many manufacturers. That is, when the color adjustment processing of a large number of manufacturers is analyzed as shown in graphs 1 to 4 and the input / output characteristics corresponding to graph 2 are extracted, an approximately S-shaped curve as shown in graph 2 is obtained. The rate of change was approximately 1.1 to 1.4 at about 50%. The rate of change was 1 in the vicinity of a gradation of 70% corresponding to the skin color luminance. Therefore, if the input / output characteristic G (sL ^* ) is determined as shown in the graph 2, it is possible to describe characteristics that are almost the same as the color adjustment processing that is uniquely determined by many manufacturers.

In graph 3, the horizontal axis represents the above-described pL ^* , and the vertical axis represents the value pY corresponding to the amount of light. In the graph 4, the horizontal axis represents the value pY corresponding to the amount of light, and the vertical axis represents the gradation value in the image data DA1. Since each input / output characteristic shown in FIG. 11 is applied to each RGB, a value can also be acquired for each RGB in the graph 4. Here, graphs 3 and 4 are input / output conversions for acquiring image data in accordance with the sRGB standard from the pL ^* . In the sRGB standard, each RGB data is acquired as data proportional to the (1 / 2.2) th power of the amount Y of light of each color. Therefore, in the graph 3, a value corresponding to the amount of light is obtained from the above-described pL ^* corresponding to lightness by the CIE Lab equation.
pY = ((pL ^* + 16) / 116) ³ (2)

Since the light amount pY after the conversion for improving the appearance of the image can be obtained by the above conversion, in the graph 4, the γ conversion is applied to the light amount pY based on the following expression. To do.
RGB = pY ^{1 / 2.2} (3)
Here, RGB is data of each color of RGB. In Expression (3), it is only necessary to calculate data proportional to the (1 / 2.2) th power of the amount of light Y, and the right side may be multiplied by a coefficient.
By generating the graph 5 of the input / output characteristics in which the input / output characteristics of the graphs 1 to 4 are superimposed by the above analysis, characteristics similar to those of the color adjustment processing in each DSC manufacturer can be described. Therefore, an information table indicating the conversion characteristic F ⁻¹ (Y) shown in the graph 5 and the reverse normal conversion characteristic Y = F (RGB) is stored in the HD 14 or the like as conversion characteristic data. As a result, the image data DA1 can be converted to a value proportional to the amount of light by the positive conversion characteristic F (RGB), and the value proportional to the amount of light can be converted by the conversion characteristic F ^-1 (Y). Can be converted to

  The printer 20 according to the present embodiment ejects each ink filled in each ink cartridge 28 provided corresponding to each color of CMYK (cyan, magenta, yellow, black) from the print heads 29a to 29d onto a print medium. It is assumed that the apparatus forms and prints ink dots. Of course, the printer 20 includes light cyan (light cyan), light magenta (light magenta), dark yellow (dark yellow), red, violet, gray (light black), light gray (light above). Printers that use non-colored ink (glossy light ink), etc., printers that do not use any of the CMYK inks, and bubble-type printers that generate bubbles in the ink path and eject ink. Such various ink jet printers, laser printers using toner ink, and the like can also be employed. Examples of the print medium include a coated paper print medium including glossy paper (for example, photographic paper) and a non-coated paper print medium including plain paper and recycled paper.

  In the printer 20, the illustrated units 21 to 27 are connected via a bus 20 a, and the CPU 21 controls each unit according to a program written in the ROM 22. The carriage that reciprocates in the main scanning direction by the carriage mechanism 27a is provided with a cartridge holder 32 in which each ink cartridge 28 is mounted, and a print head unit 29 including print heads 29a to 29d. The head drive unit 26a receives the applied voltage data corresponding to the raster data (print data) from the ASIC 26 to generate an applied voltage pattern to the piezo element, and the four color inks are applied to the print heads 29a to 29d including the element. Drop droplets in dots. The carriage mechanism 27a and the paper feed mechanism 27b connected to the I / F 27 cause the print head unit 29 to perform main scanning, or sequentially feed sheet-like print media while performing page break operations as appropriate, and perform sub-scanning. The printer 20 receives color-specific print data (raster data) from the PC 10 via the communication I / O 24 connected to the printer I / F 17e of the PC, and corresponds to the print data based on the print data. An image is formed on the print medium.

The print control device U0 shown in FIG. 2 includes image processing devices U1 to U5 and a print control unit U6, and is constructed by cooperation of a print control program including an image processing program and hardware of the PC 10. The information storage area (information recording area) U5 is provided in the HD 14, and stores the various information 14b to h described above.
The area extracting unit (area extracting unit) U1 includes units U11 and U12, and extracts a predetermined color area from multi-gradation (for example, 256 gradations) image data DA1 created by the image input device 50. The image input unit U11 inputs image data DA1 created by the image input device, and RGB data DA2 (image data) that represents a color image with a plurality of element colors RGB for each pixel (for example, 256 gradations). Type). The area acquisition unit U12 extracts a predetermined color area from the image data DA1, and generates color area information DA3 representing the range of the color area.

The color area includes a face area representing a human face area, a skin color area representing a human skin area, a sky blue area representing a sky area, a sunset color area representing a sunset area, and a green area representing a plant area. A snow-colored region representing a snow region, a color region representing a sea, lake, or river region. When the image data DA1 is data in which a color image including a person image is expressed in gradation with a plurality of element colors for each pixel, as shown in FIG. 5, the RGB data DA2 includes a color image including a person image for each pixel. The image data is expressed in gradation by color RGB. This RGB data DA2 is assumed to include the face area of the person image.
When the face area R1 is extracted from the image data DA2 shown in FIG. 5, for example, it is determined whether or not each pixel P1 constituting the image data DA2 is within the range of colors assumed as the face color. The determined pixel is extracted as a face area. For example, as shown in FIG. 4, the color range assumed as the face color is defined by a color range R11 on the ab plane of the Lab color space including the color target of the face area (target point T in the Lab color space). Can do. Of course, the color range assumed as the face color is a range defined by saturation S and hue H as a color range that is generally stored as an image by many people as a “person's skin color”. As described in Japanese Utility Model Publication No. 2004-192614, hue (H) 1YR to 9YR, which is a range of “skin tone colors” published by the Japan Color Research Institute; saturation (C) 3, 4 and 5. The coordinates of the ab plane of “preferred skin color” (target point T) are (at, bt), and a predetermined difference is Et (0 <Et <100). {(A−at) ² + (b−bt) ² } A range (a, b) on the ab plane that satisfies ^1/2 ≦ Et, a range in which the lightness L falls within a predetermined range by satisfying these conditions, and the like may be defined.
As shown in FIG. 4, a color range for extracting a color region other than the face region is also assumed as the color region R1 in the color range R11 of the color coordinate plane of the predetermined color space including the target point T of the color region R1. A range of colors can be defined. Information representing the color range R11 is stored in the information storage area U5 as color extraction information 14c. Then, referring to the color extraction information 14c, it is determined whether or not each pixel P1 constituting the image data DA2 is within the color range R11 as shown in FIG. What is necessary is just to extract as area | region R1.

  Although it is preferable that the color extraction is automatically performed when the area extraction unit is configured as described above, the configuration of the area extraction unit is specified by receiving input from the user for specifying a color range such as hue and saturation. A configuration for extracting pixels included in the color range as a color region, displaying an image corresponding to the image data, and accepting input from the user to specify the color region from the user, and specifying the pixel of the specified portion as the color region The configuration to be extracted, an image corresponding to the image data is displayed, and an input for designating an approximate image area from the image is received from the user, and whether each pixel constituting the designated image area is within the color range R11 A configuration is also conceivable in which pixels that are determined to be in the color range are extracted as color regions.

  For example, as shown in the lower part of FIG. 5, the color area information DA3 stores an integer of 1 or more in the pixel P2 determined to be within the color range R11 and 0 to the pixel P2 determined to be outside the color range R11. It can be stored data. In this case, the pixel storing 0 is extracted as the background region R2 of the color region. The color area information may be information indicating the position on the image data DA2 of the pixel determined to be in the color range R11.

となる。そして、平均値（AR,AG,AB）を求めると、ＲＧＢ色空間の色成分値Ｒ，Ｇ，ＢをＬａｂ色空間の色成分値Ｌ，ａ，ｂに変換する公知の式（６）を用いて色成分値AR,AG,ABから明度Ｌを求めることにより、色領域の明るさ（明度Lp）を取得することができる。

ただし、ｐ₁₁〜ｐ₃₃は、変換係数である。
むろん、色領域内の各画素の画素値（Ri,Gi,Bi）を先に明度Liへ変換し、平均値を求めて色領域の明るさ（明度Lp）としてもよい。また、ＲＧＢ色空間以外の色空間で定義される色成分値を用いて明るさを取得することも可能である。 The area brightness acquisition unit (area brightness acquisition unit) U2 acquires the brightness of the color area from the color area R1 extracted by the area extraction unit, and generates area brightness information DA4. The brightness of the color area can be expressed by brightness, brightness, simple average or weighted average of the color component values, etc., and is expressed by the average value, median value, mode value, etc. of the brightness of the pixels in the color area. Brightness. Therefore, the brightness of the color area can be acquired by calculating the average value, median value, mode value, etc. of the brightness of the pixels in the color area and using the area brightness information DA4. In addition, the average value of brightness values such as brightness (brightness value) includes an arithmetic average (arithmetic average) value of brightness of each pixel, a geometric average value of brightness of each pixel, The value of the harmonic average of the brightness of the pixel is conceivable, and an average value without weighting each pixel may be used, or an average value with each pixel weighted.
When the weighted average value of the brightness values is used, for example, as shown in the lower part of FIG. 5, the smallest weighting “1” is given to the pixel that touches the background region R2 in the horizontal direction or the vertical direction in the color region R1. Next, the pixel having a horizontal or vertical contact with the pixel is assigned the next smallest weighting “2”, and the pixel having the horizontal or vertical contact with the pixel is assigned a large weight one by one. The weighting is increased as the inner side of R1. The weighting value Ki (i is an integer from 1 to n, n is the number of pixels in the color area) of each pixel is stored in the RAM as color area information DA3. Here, when the pixel value of each pixel in the color area is (Ri, Gi, Bi), the weighted average value (AR, AG, AB) of the pixel values in the color area is

It becomes. Then, when the average value (AR, AG, AB) is obtained, a well-known expression (6) for converting the color component values R, G, B in the RGB color space into the color component values L, a, b in the Lab color space is obtained. By obtaining the lightness L from the color component values AR, AG, and AB, the brightness of the color region (lightness Lp ) can be obtained.

However, p ₁₁ ~p ₃₃ is a conversion coefficient.
Of course, the pixel value (Ri, Gi, Bi) of each pixel in the color area may be converted to the lightness Li first, and the average value may be obtained to obtain the brightness of the color area (lightness Lp ) . It is also possible to acquire brightness using color component values defined in a color space other than the RGB color space.

となる。そして、平均値（AR,AG,AB）を求めると、色成分値Ｒ，Ｇ，Ｂを色成分値Ｌ，ａ，ｂに変換する公知の式（６）を用いて色成分値AR,AG,ABから明度Ｌを求めることにより、背景の明るさ（明度Lq）を取得することができ。むろん、色領域の明るさを取得する場合のように、背景領域内の各画素の画素値（Ri,Gi,Bi）を先に明度Liへ変換し平均値を求めて背景領域の明るさ（明度Lq）としてもよいし、ＲＧＢ色空間以外の色空間で定義される色成分値を用いて明るさを取得することも可能である。
本実施形態では、画像データＤＡ２のうち色領域Ｒ１を除く全ての領域を背景領域Ｒ２とし、具体的には、ＲＧＢデータＤＡ２のうち顔領域Ｒ１を除く領域の全体から顔領域の背景の明るさを取得している。むろん、画像データＤＡ２のうち色領域Ｒ１を除く領域の一部のみを背景領域としてもよい。例えば、色領域の周辺のみ（色領域との境界から所定の距離内）を背景領域としたり、色領域との境界から所定の距離を超える領域を背景領域としたり、色領域以外の領域から上記色範囲Ｒ１１内の画素を除いた部分を背景領域としたりすることができる。 The background brightness acquisition unit (background brightness acquisition means) U3 acquires the brightness of the background of the color region R1 (background region R2) from the image data DA2, and generates background brightness information DA5. The brightness of the background can be represented by a simple average or weighted average of brightness, luminance, color component values, etc., and the color represented by the average value, median value, mode value, etc. of the brightness of the pixels in the background It can be. Therefore, the brightness of the background can be acquired by calculating the average value, the median value, the mode value, and the like of the brightness values of the pixels in the background area and using it as the background brightness information DA5. In addition, as the average value of the brightness value, an arithmetic average (arithmetic average) value of each pixel, a geometric average value of each pixel, a harmonic average value of each pixel, and the like are considered. An average value that does not weight the pixels may be used, or an average value that weights each pixel may be used. In the present embodiment, an arithmetic average value that does not weight each pixel value is used. That is, assuming that the pixel value of each pixel in the background area is (Ri, Gi, Bi), the arithmetic average value (AR, AG, AB) of the pixel values in the background area is

It becomes. Then, when the average values (AR, AG, AB) are obtained, the color component values AR, AG are obtained using a well-known equation (6) for converting the color component values R, G, B into the color component values L, a, b. , AB, the brightness of the background (brightness Lq ) can be obtained. Of course, as in the case of obtaining the brightness of the color area, the pixel value (Ri, Gi, Bi) of each pixel in the background area is first converted to the brightness Li and the average value is obtained to obtain the brightness of the background area ( The brightness L q) may be used, and the brightness may be acquired using a color component value defined in a color space other than the RGB color space.
In the present embodiment, all regions except the color region R1 in the image data DA2 are set as the background region R2, and specifically, the brightness of the background of the face region from the entire region excluding the face region R1 in the RGB data DA2. Is getting. Needless to say, only a part of the image data DA2 excluding the color region R1 may be used as the background region. For example, only the periphery of the color area (within a predetermined distance from the boundary with the color area) is set as the background area, the area exceeding the predetermined distance from the boundary with the color area is set as the background area, or the area other than the color area is A portion excluding pixels in the color range R11 can be used as a background region.

  Note that a region brightness information variable for generating the region brightness information DA4 and a background brightness information variable for generating the background brightness information DA5 are prepared and reset to 0, and the image data DA2 is set. The target pixel is sequentially set from among the pixels constituting the pixel, and it is determined whether or not the target pixel is in the color region R1 by determining whether or not the target pixel is in the color range R11. If it is determined that the area is a color area, the area brightness acquisition unit adds the brightness value of the pixel of interest to the area brightness information variable. If it is determined that the area is not a color area, the background brightness acquisition section Thus, the brightness value of the pixel of interest may be added to the background color region variable. When all the pixels of the image data DA2 are set, the area brightness acquisition unit divides the value of the area brightness information variable by the number of pixels in the color area to obtain area brightness information DA4, and the background brightness acquisition unit The background brightness information DA5 may be obtained by dividing the value of the background color area variable by the number of pixels in the background area.

As shown in FIGS. 1 and 13 , the adjustment unit (adjustment means) U4 changes the brightness (lightness Lp) of the color area R1 to the target brightness (lightness Lt ) preset for the color area. Is performed on the image data DA2. In the upper part of FIG. 1 shows the brightness difference obtained by subtracting the luminosity L p of the color area from the lightness Lq background ΔL of targets for (horizontal axis) lightness (vertical axis). The lower part of FIG. 1 shows an Lp-Lq plane with the brightness Lp (horizontal axis) of the color region and the brightness Lq (vertical axis) of the background as axes.
Adjustment unit U4 is the brightness of the target (lightness Lt) Dark first brightness than (lightness L1) than the brightness of the color area R1 (lightness Lp) is dark, and from the brightness of the target (lightness Lt) When the background brightness (brightness Lq ) is darker than the dark second brightness (brightness L2 ) , the brightness correction is performed on the image data DA2 with a lower brightness than the target. Further, the adjustment unit U4 has a background brightness (lightness Lq ) brighter than the brightness (lightness Lp ) of the color region R1, and the difference between the background brightness (lightness Lq ) and the brightness of the color region (lightness Lp ). Is greater than a predetermined difference, brightness correction is performed on the image data DA1 with brightness lower than the target. The difference here is, for example, that the brightness of the background and the color region is Lq and Lp, respectively, the difference Lq−Lq, the difference between the ratio Lq / Lp and 1 (Lq / Lp) −1, Lq ² −Lq ² , ( 1 / Lp) − (1 / Lq), etc.
The color space for adjusting the brightness may be a device-dependent color space, but a device-independent color space is preferable, and a uniform color space is more preferable. The uniform color space includes a CIE Lab color space, a CIE Luv color space, and the like. In addition to the uniform color space, the device independent color space includes a CIE XYZ color space, an NTS CRGB color space, and the like. Information representing the color space for adjusting the brightness is stored in the information storage area U5 as the color space information 14c. In the present embodiment, brightness is adjusted in a Lab color space including an ab plane (color coordinate plane) representing hue and saturation and a brightness axis (brightness axis). In the Luv color space, the uv plane is a color coordinate plane representing hue and saturation, and a brightness axis can be considered as the brightness axis.

The brightness of the target that represents the target of the brightness of the color area is the brightness that the person feels preferable for the color area, the brightness of the memory color stored by the person for the color area, the brightness specified by the user, etc. For example, it can be set as the center of the favorite brightness when the color is shown to a plurality of people (brightness expressed by an average value of the favorite brightness values). The target brightness can be expressed by brightness, brightness, simple average or weighted average of color component values, and the like. The target brightness of the face area can be, for example, the target brightness of the skin color, and the skin tone preference center (the brightness of the “preferred skin color” expressed by the average value of the brightness values of the favorite skin color) ). Information representing the target brightness is stored in the information storage area U5 as target brightness information 14d.
The target brightness may be constant regardless of the brightness of the color area or background, but changes according to the difference between the brightness of the background (lightness Lq ) and the brightness of the color area (lightness Lp ). By doing so, it is possible to adjust the sensuous brightness of the color area perceived by the person according to the brightness of the background, so that the brightness of the image created by the image input device can be adjusted more appropriately. It becomes possible.

The upper graph in FIG. 6 shows the sensual brightness J of the color area that is perceived by the person according to the brightness Lq of the background when the brightness of the face area is the brightness Lo of the target point T of the face area in the Lab color space. Show. Here, the sensory lightness J is lightness J specified in CIECAM02 (Colour Appearance Model 2002), which is a color appearance model published in 2004 as CIE Publication. In the upper part of the figure, the horizontal axis is the background lightness Lq, and the vertical axis is the sensory lightness J. In the example shown in the figure, the sensory brightness J of the face area having the brightness Lo becomes lower as the background brightness Lq increases within a predetermined range.
Therefore, as shown in the lower graph of the figure, the color is compensated for the difference in the sensuous brightness J of the face area that is perceived by the person according to the difference between the brightness Lq of the background and the brightness Lp of the face area. When the target brightness Lt of the area is determined, the difference in the sensory brightness of the face area due to the difference in brightness between the background and the face area is compensated, so the brightness of the image data can be adjusted more appropriately. become. In the lower part of the figure, the horizontal axis represents the background lightness Lq, and the vertical axis represents the target lightness Lt. In the example of the figure, the target brightness Lt of the face region increases as the background brightness Lq increases within a predetermined range.

  From the above, in this embodiment, as shown in FIG. 1, the lightness difference ΔL = Lq−Lp obtained by subtracting the lightness Lp of the color region from the lightness Lq of the background increases within a predetermined range (for example, ΔL1 to ΔL2). The target brightness Lt is set so as to increase continuously or stepwise. Here, when the lightness Lt continuously increases as ΔL increases, the lightness Lt may increase continuously in a monotonous manner as indicated by the line CO1 shown in FIG. 1, and the lightness Lt increases stepwise as ΔL increases. In some cases, the brightness Lt increases discretely when the lightness Lt is an integer. When the lightness Lt is an integer, the lightness Lt may be increased stepwise so as to take all integer values that the lightness Lt can take, or a part of all the integer values that the lightness Lt can take The brightness Lt may be increased stepwise so that only the integer value of is taken.

Then, the predetermined first brightness (first brightness) lower than the target brightness Lt is set as L1, and the predetermined second brightness (second brightness) lower than the target brightness Lt is set as L2. In the part U4, when the lightness Lp of the color region is lower than the lightness L1 and the lightness Lq of the background is smaller than the lightness L2, the lightness difference ΔL is determined to be a strobe non-reaching condition in which the strobe light did not reach the subject. performing a continuous or stepwise larger target lightness L t as large within brightness correction was lower in the range higher than the lightness Lp color space for the image data. In the Lp-Lq plane shown in the lower part of FIG. 1, a dark region satisfying Lp <L1 and Lq <L2 is indicated by A1. The dark area A1 is an area that is a dark image as a whole so that the image satisfies the strobe failure condition. L1 and L2 may have the same brightness or different brightness. When the image data DA2 satisfies the strobe failure condition, the adjustment unit U4 sets the brightness lower than the target Lt as Lc, and satisfies the target brightness (line CO1) in the upper part of FIG. 1 so that Lp <Lc <Lt is satisfied. ) Is lowered to the position of the line CO2, and the brightness of the image data DA2 is corrected.

Here, as indicated by the curve CU1 shown in the upper part of FIG. 7, the lightness Lc of the correction target is set according to only the lightness Lp of the color region so that Lp <Lc under the conditions of Lp <L1 and Lq <L2. Yes. The horizontal axis is Lp and the vertical axis is Lc. The lightness Lc is a function Lc (Lp) of Lp that changes according to Lp in the lightness range of 0 to L1, and information in which Lp and Lc are associated with each other is information storage area as brightness information 14h of the correction target Stored in U5. As information in which Lp and Lc are associated with each other, for example, information in a one-dimensional lookup table (LUT) format in which information defining the correspondence between Lp and Lc for each gradation can be stored. In order to satisfy Lp <Lc, for example, the function Lc (ΔL) may be set so that the minimum value of Lc is larger than Lp when Lp <L1 and Lq <L2.
The lightness Lc that is the correction target is set to lightness corresponding to only the lightness Lp of the color area because the lightness difference ΔL between the background and the color area is small in the entire image created by the image input device. Rather than setting the correction target brightness Lc accordingly, setting the correction target brightness Lc according to only the color area brightness Lp, regardless of the background brightness Lq, adjusts the image brightness appropriately. It is because it can do. On the other hand, the lightness Lc of the correction target is set lower than the lightness Lt of the target that is sensuously the same brightness. This has the following reason.
1. In a very dark image as a whole, the difference between the lightness Lp of the color area and the target lightness Lt is large, and if lightness correction is performed to increase the lightness of the color area to the target lightness Lt, noise appears in the image and the image quality after processing This is because there is a case where it does not improve as intended.
2. This is because an image captured in a very dark environment tends to make people feel less unnatural than darkness.
From the above, it can be said that the image quality after image processing can be improved by lowering the brightness correction target for an image that is very dark as a whole.

  Note that Lp <Lc is satisfied on the premise that the corrected lightness Lc is obtained by multiplying the target lightness Lt by the correction coefficient when 0 <Lp <L1 as in the curve CU2 shown in the lower part of FIG. A correction coefficient k (0 <k <1) may be set. The horizontal axis is Lp and the vertical axis is k. The correction coefficient k is a function of Lp that changes according to Lp in the range of brightness 0 to L1, and information in which Lp and k are associated is stored in the information storage area U5 as the brightness information 14h of the correction target. Is done. In order to satisfy Lp <Lc, for example, k may be set so as to satisfy k> Lp / L1.

As shown in FIG. 7, when Lp <L1 and Lq <L2, the adjusting unit U4 has a brighter color area (brightness Lp ) within a range brighter than the brightness of the color area (brightness Lp ). Brightness correction is performed on the image data DA2 by decreasing the target brightness so that it becomes brighter continuously or in steps. In other words, Lc (Lp) is Lc (Lx) ≦ Lc (Ly) (where Lc (Lx) <Lc (Ly)) using any lightness Lx, Ly satisfying 0 <Lx <Ly <L1. It is a function that at least exists. When Lc increases continuously as Lp increases, Lc (Lx) <Lc (Ly).
In other words, even if the overall image is dark, increasing the degree of lightening in a relatively dark image tends to cause noise in the processed image. Is less likely to occur. Accordingly, an image that is very dark as a whole becomes bright enough that no noise appears, and the image quality is appropriately improved.

When Lp <L1 and Lq <L2, the adjustment unit U4 lowers the target brightness within a range brighter than the brightness Lp of the color area, and has a predetermined third brightness (darker than the first brightness L1). 3rd brightness) is set to L3 (0 <L3 <L1), and brightness correction is performed on the image data DA2 so that the target brightness when Lp> L3 is higher than the target brightness when Lp <L3. You may go. That is, Lc (Lp) is a function that satisfies Lc (Lx) <Lc (Ly) using arbitrary brightness Lx and Ly that satisfy 0 <Lx <L3 <Ly <L1. As described above, noise is less likely to occur in the processed image even if it is a dark image as a whole, even if the degree of lightening is increased in a relatively bright image, so that no noise appears in a very dark image as a whole. It becomes brighter and the image quality is appropriately improved.
The third lightness L3 may be set so that 0 <L3 <L1 is satisfied, and L3 = L1 / 2 or the like may be set.

  Further, as shown in FIG. 1, the adjustment unit U4 sets a lightness difference (difference) Lq−Lp obtained by subtracting the lightness Lp of the color region from the lightness Lq of the background as ΔL, and a predetermined lightness difference larger than 0 as ΔLth. When ΔL> ΔLth, it is assumed that the backlight condition is satisfied, and brightness correction is performed on the image data DA2 by reducing the target brightness Lc according to the brightness difference ΔL. In the Lp-Lq plane shown in the lower part of FIG. 1, a backlight region that satisfies Lq-Lp <ΔLth is indicated by A2. This backlight area A2 is an area where the image satisfies the backlight condition. When the image data DA2 satisfies the backlight condition, the adjustment unit U4 sets the brightness lower than the target Lt as Lc, and sets the target brightness (line CO1) in the upper stage of FIG. 1 so that Lp <Lc <Lt is satisfied. The brightness of the image data DA2 is corrected by lowering to the position of the line CO3.

Here, the lightness Lc is a function Lc (ΔL) of ΔL that changes according to ΔL in a range where the lightness difference ΔL is larger than ΔLth, and information in which ΔL and Lc are associated with each other is brightness information of the correction target. 14h is stored in the information storage area U5. The information in which ΔL and Lc are associated with each other can be, for example, information in a one-dimensional lookup table (LUT) format that stores information that defines the correspondence between ΔL and Lc for each gradation.
The reason why the lightness Lc of the correction target is lower than the lightness Lt of the target that is sensuously the same brightness is as follows.
That is, in an image created under backlight conditions by an image input device, the brightness Lp of the color area is low, and as a result, the difference between the brightness Lp of the color area and the target brightness Lt is large, and the brightness of the color area is set to the target brightness. This is because when the brightness correction is increased to Lt, noise appears in the image and the processed image quality may not be improved as intended.
Therefore, in an image created under backlight conditions, it can be said that lowering the brightness correction target can improve the image quality after image processing.

Further, as shown in FIG. 1, the lightness Lc of the correction target is set according to the lightness difference ΔL so that Lp <Lc under the condition of Lq−Lp> ΔLth. In order to satisfy Lp <Lc, for example, the function Lc (ΔL) may be set so that the minimum value of Lc is larger than Lp when Lq−Lp> ΔLth. When ΔL> ΔLth, the adjustment unit U4 performs brightness correction on the image data DA2 by decreasing the target brightness Lc continuously or stepwise as the brightness difference ΔL increases. That is, Lc (ΔL) is at least Lc (ΔLx) ≧ Lc (ΔLy) (where Lc (ΔLx)> Lc (ΔLy) using an arbitrary brightness difference ΔLx, ΔLy that satisfies ΔLth <ΔLx <ΔLy. Exist)). When Lc decreases continuously as ΔL increases, Lc (ΔLx)> Lc (ΔLy).
That is, even in an image created under backlight conditions, if the degree of brightening is increased as a result of a relatively dark color area in an image with a relatively large degree of backlighting, noise tends to occur in the processed image, while the degree of backlighting is relatively high. In a small image, noise is less likely to occur in the processed image even if the degree of brightening is increased as a result of the relatively bright color area. Therefore, the image created under the backlight condition is brightened to the extent that noise does not appear, and the image quality is appropriately improved.

As shown in FIG. 1, when ΔL> ΔLth, the adjustment unit U4 lowers the target brightness within a range brighter than the brightness Lp of the color area, and at the same time a predetermined second greater than the brightness difference ΔLth. Lightness correction may be performed on the image data DA2 by setting the lightness difference to ΔL3 (ΔLth <ΔL3 <ΔL2) and setting the target lightness when ΔL> ΔL3 to be lower than the target lightness when ΔL <ΔL3. . That is, Lc (ΔLp) is a function that satisfies Lc (ΔLx)> Lc (ΔLy) using arbitrary brightness differences ΔLx and ΔLy that satisfy ΔLth <ΔLx <ΔL3 <ΔLy <ΔL2. As described above, even in an image created under backlight conditions, noise is less likely to occur in the processed image even if the degree of brightening is increased as a result of the relatively bright color area in an image with relatively small backlighting. The image created under the backlight condition becomes bright enough that no noise appears, and the image quality is appropriately improved.
The second lightness difference ΔL3 may be set so that ΔLth <ΔL3 <ΔL2 is satisfied, and may be ΔL3 = (ΔLth + ΔL2) / 2 or the like.

  The adjustment unit according to the present embodiment performs brightness correction uniformly on the entire image data, but may perform brightness correction only on the extracted color region. In the latter case, when the correction characteristics (e.g., ε described later) of the pixel value of the RGB data DA2 are determined, the target pixel is sequentially set from each pixel constituting the color area, and the correction characteristics are referred to. The converted RGB data DA6 can be generated by converting the pixel value of the pixel of interest according to the correction characteristic information.

  When the final target lightness Lp ′ in the Lab color space is determined, the RGB data DA6 after adjustment is performed by performing lightness correction according to the difference (ratio) between the lightness Lp and the lightness Lp ′ of the color area before correction. (A kind of image data) is generated.

As described above, for a very dark image as a whole, the image quality can be improved by lowering the brightness correction target in the dark direction. In the present invention, when Lp <L1 and Lq <L2, the target brightness is lowered, so that it is possible to solve the problem of noise in the image. Therefore, even when the appropriate brightness adjustment is not performed in the image input device, it is possible to appropriately adjust the brightness of the image data created by the image input device.
Even in an image created under backlight conditions, the image quality can be improved by lowering the brightness correction target in the dark direction. In the present invention, when ΔL> ΔLth, the target brightness is lowered, so that it is possible to solve the problem of noise in the image. Therefore, even when the appropriate brightness adjustment is not performed in the image input device, it is possible to appropriately adjust the brightness of the image data created by the image input device.

ただし、ｑ₁₁〜ｑ₃₃は、変換係数である。
そして、調整前のＲＧＢデータの全画素の画素値に対して補正前後の色成分値Ｒ，Ｇ，Ｂの差異（比）に応じた補正を行うことにより、色バランス調整後のＲＧＢデータを生成することができる。 Note that the adjustment unit U4 may further perform color balance adjustment on the image data DA2 so that the color of the color region R1 approaches the preset target color Ct for the color region R1. For example, the representative color Cp of the color area is extracted from the color area R1 extracted by the area extraction unit U1, the background representative color Cq of the color area R1 is extracted from the image data DA2 as necessary, and the target color Ct. When the image data DA2 is subjected to a color balance adjustment that brings the color of the color region R1 closer to the target color Ct based on the representative color Cp and, if necessary, the representative color Cq, the color balance of the image as well as the brightness of the image Adjusted. The average value (AR, AG, AB) of the pixel values (Ri, Gi, Bi) of each pixel in the color area and the background area is obtained by the above formula (4) and the above formula (6), and Lab is obtained by the above formula (5). When converted to the color component values (AL, Aa, Ab) of the color space, the representative colors Cp, Cq of the color area and background area are extracted as the color component values (AL, Aa, Ab). Therefore, using the extraction result and the color component value of the target color Ct, the position P (a _p , a _p ) of the representative color Cp in the color region on the ab plane is brought close to the position T (a _t , b _t ) of the target color Ct. The point P ′ (a _{p ′} , b _{p ′} ) can be obtained. For example, when the correction coefficient a ratio closer the point P to the point T α (0 <α ≦ 1 ), a p '= a p + α (a t -a p), b p' = b p + α (b t - b _p ). The correction coefficient α may be changed according to the position of the background representative color Cq. When the color component values (AL, a _{p ′} , b _{p ′} ) after correction in the Lab color space are determined, the color component values L, a, b in the Lab color space are converted into the color component values R, G, R, in the RGB color space. The corrected color component values L, a, and b are converted into color component values R, G, and B using a well-known equation (7) that converts the color components into B.

However, q ₁₁ to q ₃₃ is the conversion factor.
Then, the RGB data after the color balance adjustment is generated by performing correction according to the difference (ratio) of the color component values R, G, and B before and after the correction on the pixel values of all the pixels of the RGB data before adjustment. can do.

The print control unit U6 performs color conversion on the RGB data DA6 expressed with the element color RGB for each pixel into CMYK data expressed with the element color CMYK for each pixel (for example, expressed with 256 gradations), and the color conversion is performed. A predetermined halftone process is performed on the subsequent CMYK data to reduce the number of gradations to m gradations (m is an integer of 2 or more and less than 256), and a predetermined rasterization process is performed on the CMYK data after the halftone process. Then, raster data (print data) DA7 is generated, and processing for outputting the raster data DA7 to the printer 20 is performed. This process is a process for controlling the printer to form ink dots corresponding to the CMYK data on the print medium based on the CMYK data. By this processing, control for causing the printer to print an image corresponding to the RGB data based on the RGB data DA6 is performed.
Then, the printer 20 forms an image corresponding to the RGB data DA6 on the print medium M1 by attaching ink dots corresponding to the CMYK data on the print medium M1. As a result, an image IM1 with appropriate brightness is printed on the print medium.

(2) Image processing operations and effects:
When the print control process shown in FIG. 8 is started, image data DA1 created by the image input device is input by the image input unit, and converted into, for example, RGB data DA2 expressed in a wide RGB color space (step S102, hereinafter). , “Step” is omitted). The image data DA1 can be input directly from the image input device 50, image data stored in the DSC memory card 54i can be input via the memory card I / F, or the image input device. The image data can also be input from the CD-ROM 15a or the like on which the image data created in step 1 is recorded. When inputting image data, not only the entire image data is stored in the RAM, but the image data is divided and the divided data may be overwritten in the RAM sequentially, or a buffer area storing the image data is stored. Only represented data may be stored in the RAM. The image data DA1 is data in which an image is expressed in gradation by a large number of pixels in a dot matrix, and is image data composed of RGB defined in the sRGB color space or an image composed of YUV in the YUV color system. There are data, etc. Since each component of the image data DA1 has various gradation numbers, the image data DA1 is converted into, for example, 256 gradations of R, G, and B (integer values of 0 to 255) according to the definition of sRGB, YUV color system, etc. ) RGB data DA2.

  Next, the area acquisition unit performs a process of extracting a color area R1 such as a predetermined face area from the RGB data DA2 (S104). For example, an area for creating the color area information DA3 is secured in the RAM, the color space information 14b and the color extraction information 14c are read from the HD to the RAM, and the target pixel is sequentially set from each pixel of the image data DA2. Then, it is determined whether or not the target pixel is within the color range R11 with reference to the color extraction information 14c, and the target pixel is within the color range at a location corresponding to the target pixel in the region to be the color region information DA3. 1 is stored, and when it is determined that the target pixel is not within the color range, 0 is stored. Thereby, color area information DA3 representing the range of the color area is generated.

Further, the brightness Lp of the color area is acquired from the color area R1 by the area brightness acquisition unit (S106). For example, the pixel values (Ri, Gi, Bi) of the pixels corresponding to the locations where 1 or more are stored in the color area information DA3 among the pixels of the RGB data DA2 are arithmetically averaged without weighting, and the average value (AR, AG, AB) is calculated and converted to color component values L _p , a _p , b _p representing the representative colors of the color area using the above formula (5), and the brightness L _p is set as area brightness information DA4. Alternatively, as shown in the lower part of FIG. 5, processing for generating color area information DA3 having a larger weight Ki toward the inner side of the color area R1 is performed, and among the pixels of the RGB data DA2, one or more of the color area information DA3 is one or more. The weighted Ki is given to the pixel value (Ri, Gi, Bi) of the pixel corresponding to the stored location, and the weighted average value (AR, AG, AB) is calculated by arithmetic averaging according to the above equation (4). The component values L _p , a _p , and b _p are converted, and the lightness L _p is set as area brightness information DA4.

  Thereafter, the background brightness acquisition unit acquires the brightness Lq of the background of the color region R1 from the RGB data DA2 (S108). For example, the pixel value (Ri, Gi, Bi) of the pixel corresponding to the place where 0 is stored in the color area information DA3 among the pixels of the RGB data DA2 is arithmetically averaged by the above formula (6) without weighting. An average value (AR, AG, AB) is calculated, converted to brightness AR using the above equation (5), and this brightness is used as background brightness information DA5.

As will be described in detail later, the adjustment unit refers to the target brightness information 14d, the area brightness information DA4, the background brightness information DA5, the threshold values 14e to f, and the corrected target brightness information 14h. Brightness correction is performed on the RGB data DA2 to set the brightness of the region to the target brightness, and adjusted RGB data DA6 is generated (S110).
The PC 10 that performs the processes of S102 to S110 constitutes an image processing apparatus.

  After that, the print control unit refers to the color conversion LUT while sequentially moving the target pixel to be color-converted among the pixels of the RGB data DA6, thereby determining the R, G, and B gradation values of the target pixel. Conversion to C, M, Y, and K gradation values generates CMYK data (S112). The color conversion LUT is an input data that expresses the color before color conversion with gradation values for each of the element colors RGB, and an output that expresses a color that is reproduced with CMYK ink on the print medium with gradation values for each CMYK. It is an information table that defines a correspondence relationship with data for a plurality of reference points. The CMYK data is image data in which an image is expressed in gradation by a large number of pixels arranged in a dot matrix like the RGB data DA6. The gradation data for each pixel is the usage amount of each ink ejected from the print head by the printer. Suppose that the data is 256 gradations of C, M, Y, and K.

Next, predetermined halftone processing such as an error diffusion method, a dither method, and a density pattern method is performed on the CMYK data to reduce the number of gradations, and multi-value data for each CMYK is generated (S114). The multi-value data is data representing the dot formation status as the presence / absence of dot formation. For example, the gradation value “1” is binarized by corresponding to dot formation and the gradation value “0” corresponding to no dot formation. Two-level binary data can be obtained. Of course, it is good also as data of 4 gradations. Further, predetermined rasterization processing (interlace processing) is performed on the multi-value data to rearrange the pixels in the order used by the printer, and raster data (print data) DA7 for each CMYK is generated (S116). Then, the raster data DA7 is output to the printer 20 (S118), and the print control process is terminated. As a result, the PC controls the printer to form dots corresponding to the multi-value data and print the image.
Then, the printer 20 obtains the raster data DA7, forms each dot of CMYK ink corresponding to the CMYK data based on the raster data, and outputs the image IM1 corresponding to the RGB data DA6 after color adjustment to the print medium M1. Print up.

FIG. 9 is a flowchart showing the adjustment process performed in S110. When the process is started, first, the target brightness Lt of the color region R1 is acquired (S202). For example, the target brightness information 14d defining the correspondence between the brightness difference ΔL = Lq−Lp and the target brightness Lt shown in FIG. 1 is read from the HD to the RAM, and the variable LLast for storing the final target brightness is stored. Is secured in the RAM, the target brightness Lt (ΔL) corresponding to the brightness difference ΔL is obtained, and the value of this Lt (ΔL) is substituted into the variable LLast. Of course, the target brightness may be acquired by receiving an input specifying the target brightness from the user and acquiring information indicating the specified brightness as the target brightness information.
Next, it is determined whether or not the lightness Lp of the color region is smaller than the first lightness L1 and the lightness Lq of the background is smaller than the second lightness L2 (S204). When the condition is satisfied, assuming that the image data DA2 satisfies a predetermined strobe non-achieving condition (strobe non-achieving mode), refer to the correction target brightness information 14h that defines the correspondence between the lightness Lp and the lightness Lc shown in FIG. Then, the correction target lightness Lc (Lp <Lc <Lt) lowered from the target Lt according to only the lightness Lp of the color region is set (S206), and the process proceeds to S212. The value of Lc is substituted for the variable LLast. Note that if Lp ≧ L1 or Lq ≧ L2 and not the backlight condition, the background is extremely bright and the entire image is bright to some extent, so the lightness of the normal mode that makes the lightness of the color area the target lightness Lt No noise appears in the image even after correction.

When the condition is not satisfied in S204, it is determined whether or not the lightness difference ΔL = Lq−Lp obtained by subtracting the lightness Lp of the color area from the lightness Lq of the background is smaller than a predetermined lightness difference ΔLth (S208). When the condition is satisfied, assuming that the image data DA2 satisfies a predetermined backlight condition (backlight mode), referring to the correction target brightness information 14h that defines the correspondence between the brightness difference ΔL and the brightness Lc shown in FIG. The correction target brightness Lc (Lp <Lc <Lt) lowered from the target Lt according to the brightness difference ΔL is set (S210), and the process proceeds to S212. The value of Lc is substituted for the variable LLast. Note that if Lp ≧ L1 or Lq ≧ L2 and the backlight condition is not met, the color area will not be too dark as a result of the extremely bright background, so the brightness of the color area is set to the target brightness Lt. Noise does not appear in the image even after brightness correction.
When the conditions are not satisfied in S204 and S208, it is determined that the normal mode (predetermined normal imaging conditions) is set, and the process proceeds to S212 without decreasing the target brightness Lt.

In S212, the correction magnification ε for the pixel value of each pixel of the RGB data DA2 is determined using the value Lp ′ of the variable LLast and the brightness Lp of the color area. Then, the pixel value of the RGB data DA2 is corrected for each pixel based on the correction magnification ε, the adjusted RGB data DA6 is generated (S214), and the adjustment process is terminated.
Through the above processing, brightness correction is performed on the image data DA2 so that the brightness of the color area becomes the target brightness.

In FIG. 10, the lower graph shows the image quality adjustment characteristics when brightness correction (exposure correction) is performed, and the upper graph shows the characteristics of the conversion process to be performed to obtain the image quality adjustment characteristics. ing. In these graphs, the horizontal axis represents an input value and the vertical axis represents an output value, which indicates that each input value is converted into a corresponding output value on the graph according to the characteristics indicated by the solid line. A straight line indicated by a broken line indicates a characteristic when the input value is equal to the output value. Graph 6 is a positive conversion characteristic Y = F (RGB), graph 8 is a conversion characteristic F ⁻¹ (Y), and graph 7 is an input / output characteristic for exposure correction. According to the positive conversion characteristic shown in the graph 6, the pixel value of the RGB data DA2 can be converted into the amount of light Y for each RGB color component. Since the exposure correction can be performed by multiplying the amount of light by a constant, the exposure correction characteristic is defined by multiplying the amount of light Y by ε as shown in the graph 7. Here, if the brightness of the color area before and after adjustment is Lp and Lp ′, respectively,
ε = Lp '/ Lp (7)
It is said.
In FIG. 10, the amount of light after exposure correction is represented as eY, the horizontal axis represents the amount of light Y, and the vertical axis represents the amount of light eY after exposure correction. If the light amount eY after exposure correction is converted with the conversion characteristic F ⁻¹ as shown in the graph 8, image data after exposure correction can be acquired for each RGB color component.

A graph 9 is a result of superimposing the conversion characteristics of the graphs 6 to 8 and is the image quality adjustment characteristic. That is, the horizontal axis represents the RGB color component values in the image data DA2, and the vertical axis represents the image data DA6 after exposure correction has been performed for the RGB color components.
Therefore, the correspondence relationship between the R, G, and B pixel values in the RGB data DA2 before adjustment and the R, G, and B pixel values in the adjusted RGB data DA6 according to the value of the correction magnification ε for each gradation. The specified LUT (information table) is stored in the HD 14 or the like as correction correspondence information, and the RGB values DA2 before adjustment are converted for each element color RGB by referring to this LUT, thereby adjusting the adjusted RGB. Data DA6 can be generated.
The conversion according to the positive conversion characteristic Y = F (RGB) and the conversion according to the conversion characteristic F ⁻¹ (Y) are preferable in that the image after adjustment can have a very high image quality. In order to simplify the process, the R, G, B pixel values of the RGB data DA2 are multiplied by the correction magnification ε without performing conversion according to Y = F (RGB) and F ⁻¹ (Y). The RGB data DA6 may be generated. Also, an LUT that defines the correspondence between the pixel value before adjustment and the pixel value after adjustment according to the value of the correction magnification ε is prepared for each pixel and stored in the HD 14 or the like, and the value of the correction magnification ε is set. The RGB data DA6 may be generated by converting the R, G, and B pixel values of the RGB data DA2 according to the information of the LUT with reference to the corresponding LUT.

Note that color balance adjustment may be performed on the RGB data DA6 after brightness correction. As described above, in the Lab color space, the target point T (L _t , a _t , b _t ) of the color region and the point P (L _p , a _p , b _p ) of the representative color Cp of the color region are corrected. the point _{P '(L p, a p} + α (a t -a p), b p + α (b t -b p)) of the calculated, the equation (7) color component values AR in the RGB color space using a' By converting to AG ′, AB ′ and determining the correction magnifications γ, δ, β for the element colors RGB, the pixel values of the RGB data DA6 can be corrected for each pixel based on the correction magnifications γ, δ, β. . For example, if a LUT that defines the correspondence between the pixel value before adjustment and the pixel value after adjustment according to the correction magnifications γ, δ, β is prepared for each pixel and stored in the HD 14 or the like, the correction magnification γ R, G, B pixel values of the RGB data DA6 are converted according to the information of the LUT with reference to the LUT corresponding to the values of, δ, β, and the RGB data after color balance adjustment can be generated.
Of course, the brightness correction of the present invention may be performed after the color balance adjustment is performed on the RGB data DA2 before the brightness correction.
As described above, the color balance of the image is also adjusted, so that the image quality is further improved.

  When the adjustment process is completed, control is performed to cause the printer 20 to print an image corresponding to the RGB data DA6 after adjustment in S112 to S118 of FIG. Then, the printer prints the image IM1 adjusted corresponding to the RGB data DA6 on the print medium M1.

FIG. 13 schematically shows brightness correction of each image in which the color region and the background have various brightnesses. Here, the horizontal axis represents the difference in brightness between the background and the color area, and is represented by the above-described brightness difference ΔL = Lq−Lp or the like. The vertical axis represents the brightness of the color area, which is represented by the above-described lightness Lp and the like, and represents the exposure. If it is not a strobe failure condition or a backlight condition, the brightness is corrected so that the color area such as the face area has the same brightness sensuously as the normal condition (normal mode). That is, under normal conditions, when the brightness of the color area is above the line CO1, the brightness of the color area is lowered to the position of the line CO1, and when the brightness of the color area is below the line CO1. The brightness of the color area is increased to the position of line CO1.
Here, under the condition that Lp <L1 and Lq <L2, that is, the strobe failure condition, the lightness of the color area is on the lower side of the line CO1, and the lightness of the color area is raised, but it is below the line CO1. It is stopped at the brightness. Thereby, the image is brightened to the extent that noise does not occur, and the image quality can be improved appropriately.
In addition, under the condition of Lq−Lp> ΔLth, that is, the backlight condition, the lightness of the color region is assumed to be lower than the line CO1, and the lightness of the color region is raised but stopped at the lightness below the line CO1. . Thereby, the image is brightened to the extent that noise does not occur, and the image quality can be improved appropriately.

As described above, according to the present invention, even when a very dark image is created entirely by the image input device, such as when the strobe light does not reach the subject, the target brightness is reduced. Since it can be lowered appropriately, it is possible to prevent image corruption such as noise in the image. Further, even when an image input device creates an image with a dark color area under backlight conditions, the target brightness can be appropriately lowered, so that it is possible to prevent image corruption such as noise in the image. Therefore, it is possible to provide an image processing apparatus capable of performing appropriate brightness correction (exposure correction) at the same time under the normal condition, the strobe failure condition, and the backlight condition, and the appropriate brightness adjustment cannot be performed by the image input apparatus. Even in such a case, it is possible to appropriately adjust the brightness of the image created by the image input device.
Of course, the same can be said for adjusting the brightness of an image having a color area other than the face area.

(3) Modification:
The printing system including the image processing apparatus of the present invention can have various configurations. For example, the printing apparatus may be an apparatus integrated with a computer or a printing apparatus that prints only a single color image. Some or all of the above-described flows may be executed by a printing apparatus, a dedicated image processing apparatus, an external server, or the like. If the printing apparatus is an apparatus that can execute halftone processing and rasterization processing, the image data after color conversion may be output to the printing apparatus as it is.
In addition to RGB data, image data to be corrected for brightness includes YUV data composed of pixel values for each color component YUV defined in the YUV color system, Lab data composed of pixel values for each color component Lab, and element colors CMY. CMY data including pixel values for each pixel, CMYK data including pixel values for each element color CMYK, and the like may be used.
Although it is preferable that the target brightness is set so as to compensate for a difference in sensory brightness such as the brightness J, even if the target brightness is constant regardless of the background brightness, An effect of appropriately improving brightness without causing noise in the image under unreached conditions or backlight conditions can be obtained.

The first brightness (lightness L1) and the second brightness (lightness L2) for determining the dark area A1 depend on the brightness of the color area and the background (lightness Lp, Lq) in addition to the fixed value. It may be a variable value that changes. For example, L1 = k ₁ × Lt (0 <k ₁ <1, for example, k ₁ = 0.5), L2 = k ₂ × Lt (0 <k ₂ <1, for example, k ₂ = 0.5), and target L1 and L2 The brightness can be changed according to the brightness (lightness Lt). Also, L1 = L1′−k ₃ × Lp (0 <L1 ′ <Lt, 0 <k ₃ <1), L2 = L2′−k ₄ × Lq (0 <L2 ′ <Lt, 0 <k ₄ <1 ).
The predetermined difference in brightness (lightness difference ΔLth) for determining the backlight region A2 may be a variable value that changes according to the brightness of the color region or the background (lightness Lp, Lq), in addition to a fixed value. For example, ΔLth = ΔLth ′ + k ₅ × Lp (−1 <k ₅ <1), ΔLth = ΔLth ′ + k ₆ × Lq (−1 <k ₆ <1), ΔLth = ΔLth ′ + k ₇ × (Lp + Lq) (− 1 <k ₇ <1), and ΔLth can be set to a lightness that changes in accordance with the brightness of the color region and the background (lightness Lp, Lq).

Since the target brightness Lt is the brightness according to ΔL = Lq−Lp and the correction target brightness Lc under the backlight condition is also the brightness according to ΔL, the processing of S208 to S210 in FIG. 9 can be omitted. .
FIG. 14 is a flowchart showing the adjustment process performed by the image processing apparatus according to the modification. In this modification, the target brightness Lt when ΔL ≦ ΔLth, and the brightness of the correction target that defines the correspondence between the correction target brightness Lb and ΔL that becomes the correction target brightness Lc when ΔL> ΔLth. Information is prepared and stored in the HD 14. First, instead of S202, the correction target brightness (brightness) Lb of the color region R1 is acquired (S202). Here, the brightness Lb of the correction target corresponding to the brightness difference ΔL is obtained with reference to the brightness information of the correction target. Next, it is determined whether Lp <L1 and Lq <L2 are satisfied (S204). When the condition is satisfied, the correction target brightness information 14h defining the correspondence between the lightness Lp and the lightness Lc shown in FIG. 7 is referred to, and the correction target lowered from the target Lt according to only the lightness Lp of the color region The brightness Lc (Lp <Lc <Lt) is set (S206), and the process proceeds to S212. On the other hand, when the condition is not satisfied, the process immediately proceeds to S212. In S212, the correction magnification ε is determined using the value Lp ′ of the variable LLast and the brightness Lp of the color area (S212). Then, the adjusted RGB data DA6 is generated from the RGB data DA2 based on the correction magnification ε (S214).
Also in this process, the brightness of the image is appropriately adjusted under the strobe failure condition and the backlight condition.

In addition, when the brightness of the color area falls within the predetermined range, brightness correction is not performed assuming that the brightness is within the range of “preferred brightness”, and the brightness of the color area falls outside the predetermined range. The adjustment unit may be configured to perform only brightness correction.
FIG. 15 is a flowchart illustrating print control processing performed by the print control apparatus according to the modification. In this process, S130 is inserted between S106 and S108. That is, when the brightness (lightness Lp) of the color region is acquired in S106, it is determined in S130 whether the lightness Lp is within a predetermined range L _{L to} L _U of “preferable brightness”. However, 0 <L _L <Lt <L _U <100. When the lightness Lp is between the boundaries L _L and L _U of the predetermined range, it may be determined that it is within the range or may be determined to be outside the range. When it is determined that the lightness Lp is outside the range, the processing of S108 to S118 is performed. On the other hand, when the lightness Lp is determined to be within the range, the process proceeds to S112 and the adjustment processing is omitted. As a result, if the brightness of the color region is within the range of “preferable brightness”, the process of adjusting the brightness of the image data is omitted, so that the image processing can be speeded up.

Incidentally, even the image processing apparatus that does not include the print control unit U6 shown in FIG. 2 can obtain the basic functions and effects of the present invention. Further, even in an image processing apparatus that does not include the image input unit U11 in the region extraction unit U1, the basic operation and effect of the present invention can be obtained.
In addition, this invention is not restricted to the Example and modification which were mentioned above, Each structure disclosed in the Example and modification which were mentioned above mutually replaced, the structure which changed the combination, well-known technique, and the above-mentioned Configurations in which the respective configurations disclosed in the embodiments and the modified examples are mutually replaced or combinations are changed are also included.
As described above, according to the present invention, the brightness of the image data created by the image input device is appropriately adjusted according to various aspects even when the image input device has not performed appropriate brightness correction. It becomes possible.

The figure which shows typically the image processing which is one Embodiment of this invention. FIG. 2 is a block diagram schematically illustrating a configuration of a print control apparatus including an image processing apparatus. 1 is a block diagram showing an outline of a configuration of a printing system including a printing control apparatus. The figure which shows typically the color range made into a color area | region in ab plane. The figure which shows a mode that a color area | region is extracted from image data. The figure which shows typically a mode that the sensory brightness J which changes according to the brightness of a background is compensated. The figure which shows typically the correction | amendment target of the brightness of the color area in a dark image. 6 is a flowchart illustrating print control processing performed by the print control apparatus. 6 is a flowchart showing adjustment processing performed by the image processing apparatus. The figure which shows a mode that the brightness of image data is correct | amended. The figure which shows typically the color adjustment process which a digital still camera performs, and its conversion characteristic. The block diagram which shows the structure of a digital still camera typically. The figure which shows the effect | action of this invention typically. 9 is a flowchart illustrating adjustment processing performed by a print control apparatus according to a modification. 9 is a flowchart illustrating print control processing performed by a print control apparatus according to a modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Personal computer (PC), 14 ... Hard disk (HD), 14b ... Color space information, 14c ... Color extraction information, 14d ... Target brightness information, 14e ... First brightness (first brightness), 14f ... second brightness (second brightness), 14g ... predetermined brightness difference (predetermined difference), 14h ... corrected target brightness information with reduced brightness, 20 ... inkjet printer (printing apparatus), 50 ... Digital still camera (image input device), A1 ... dark area, A2 ... backlight area, DA1, DA2 ... image data, DA3 ... color area information, DA4 ... area brightness information, DA5 ... background brightness information, DA6 ... after adjustment Image data, R1... Color region, R2... Background region, R11... Color range, U0... Print control device, U1... Region extraction unit (region extraction means), U2. ), U3 ... background brightness acquisition unit (background brightness acquisition unit), U4 ... adjustment unit (adjustment unit), U5 ... information storage area (information recording area), U6 ... print control unit, U11 ... image input unit, U12 ... area acquisition unit,

Claims (14)

Area extracting means for extracting a predetermined color area from image data created by the image input device; area brightness acquiring means for acquiring brightness of the color area from the extracted color area; and brightness of the color area. An image processing apparatus comprising: an adjustment unit that performs brightness correction on the image data so as to obtain a target brightness set in advance for the color region;
Background brightness acquisition means for acquiring the brightness of the background of the color region from the image data;
The adjusting means, when the brightness of the color area is darker than the first brightness darker than the target brightness, and when the background brightness is darker than the second brightness darker than the target brightness, An image processing apparatus that performs brightness correction on the image data with brightness lower than a target. The target brightness is a brightness that changes according to the difference between the brightness of the background and the brightness of the color area,
When the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the adjustment means is configured to make a difference between the brightness of the background and the brightness of the color area. The image processing apparatus according to claim 1, wherein brightness correction is performed on the image data by reducing a target brightness that varies according to the brightness within a range brighter than the brightness of the color region. All the brightness is expressed in brightness,
The target brightness is a brightness that increases continuously or stepwise as the brightness difference ΔL obtained by subtracting the brightness of the color area from the brightness of the background increases within a predetermined range.
When the lightness of the color area is lower than the first lightness lower than the target lightness and the lightness of the background is lower than the second lightness lower than the target lightness, the adjustment means The brightness correction is performed on the image data so that a target brightness that increases continuously or stepwise as ΔL increases within a predetermined range is lowered within a range higher than the brightness of the color region. The image processing apparatus according to claim 1 or 2 . When the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the adjusting means may adjust the color within a range brighter than the brightness of the color area. any of claims 1 to 3, characterized in that performing the brightness correction for lowering the brightness of the target as the brightness of the region is higher becomes continuously or stepwise bright bright on the image data the image processing apparatus according to an item or. When the brightness of the color area is darker than the first brightness and the brightness of the background is darker than the second brightness, the adjusting means sets the target brightness to be higher than the brightness of the color area. The target brightness when the brightness of the color area is lower than the third brightness that is darker than the first brightness is reduced within the bright range, and the brightness of the color area is darker than the third brightness 4. The image processing apparatus according to claim 1 , wherein brightness correction that is brighter than the target brightness at the time is performed on the image data. 5. The adjustment means further reduces the brightness below the target when the brightness of the background is brighter than the brightness of the color area and the difference between the brightness of the background and the brightness of the color area is greater than a predetermined difference. 6. The image processing apparatus according to claim 1 , wherein brightness correction is performed on the image data. The adjusting means has a brightness that is brighter than the brightness of the color area when the brightness of the background is brighter than the brightness of the color area and the difference between the brightness of the background and the brightness of the color area is greater than a predetermined difference. in claim and performs the brightness and the difference becomes large enough to continuously or stepwise brightness correction was lowered brightness of the target with the brightness of the color area of the background to the image data 6 An image processing apparatus according to 1. All the brightness is expressed in brightness,
The target brightness is a brightness that increases continuously or stepwise as the brightness difference ΔL obtained by subtracting the brightness of the color area from the brightness of the background increases.
When the ΔL is larger than a predetermined lightness difference ΔLth (ΔLth> 0), the adjustment means continuously or stepwise increases the lightness of the target as the lightness difference ΔL increases within a range higher than the lightness of the color area. The image processing apparatus according to claim 7 , wherein the image data is subjected to brightness correction with a low value. The image data is data in which a color image including a human image is expressed in gradation with a plurality of element colors for each pixel,
The predetermined color area is a face area of the person image,
The target brightness set in advance for the face area is the brightness of the skin color target, and the brightness that increases continuously or stepwise as the brightness difference ΔL increases when the ΔL is equal to or less than the ΔLth. And
The background brightness acquisition means acquires the brightness of the background of the face area from the entire area of the image data excluding the face area,
When the lightness of the color area is lower than the first lightness lower than the target lightness and the lightness of the background is lower than the second lightness lower than the target lightness, the adjustment means continuously or stepwise larger target brightness as ΔL increases to lower in a higher range than the brightness of the color area, according to claim 8, wherein the performing brightness correction of the image data Image processing device. Area extracting means for extracting a predetermined color area from image data created by the image input device; area brightness acquiring means for acquiring brightness of the color area from the extracted color area; and brightness of the color area. An image processing apparatus comprising: an adjustment unit that performs brightness correction on the image data so as to obtain a target brightness set in advance for the color region;
Background brightness acquisition means for acquiring the brightness of the background of the color region from the image data;
The adjusting means has a brightness lower than the target when the brightness of the background is brighter than the brightness of the color area and the difference between the brightness of the background and the brightness of the color area is larger than a predetermined difference. An image processing apparatus, wherein the image correction is performed on the image data. A predetermined color area is extracted from the image data created by the image input device, the brightness of the color area is acquired from the extracted color area, and the brightness of the color area is set to a target value set in advance for the color area. An image processing method for performing brightness correction for brightness on the image data,
The brightness of the background of the color area is acquired from the image data, the brightness of the color area is darker than the first brightness that is darker than the target brightness, and the second brightness is darker than the target brightness. When the brightness of the background is darker, the image processing method is characterized in that brightness correction is performed on the image data with brightness lower than the target. A predetermined color area is extracted from the image data created by the image input device, the brightness of the color area is acquired from the extracted color area, and the brightness of the color area is set to a target value set in advance for the color area. An image processing method for performing brightness correction for brightness on the image data,
When the brightness of the background of the color area is acquired from the image data, the brightness of the background is brighter than the brightness of the color area, and the difference between the brightness of the background and the brightness of the color area is greater than a predetermined difference An image processing method comprising performing brightness correction on the image data with brightness lower than the target. A predetermined color area is extracted from the image data created by the image input device, the brightness of the color area is acquired from the extracted color area, and the brightness of the color area is set to a target value set in advance for the color area. An image processing program for causing a computer to realize a function of performing brightness correction for brightness on the image data,
The brightness of the background of the color area is acquired from the image data, the brightness of the color area is darker than the first brightness that is darker than the target brightness, and the second brightness is darker than the target brightness. An image processing program for realizing a function of performing brightness correction on the image data with brightness lower than the target when the background is dark. A predetermined color area is extracted from the image data created by the image input device, the brightness of the color area is acquired from the extracted color area, and the brightness of the color area is set to a target value set in advance for the color area. An image processing program for causing a computer to realize a function of performing brightness correction for brightness on the image data,
When the brightness of the background of the color area is acquired from the image data, the brightness of the background is brighter than the brightness of the color area, and the difference between the brightness of the background and the brightness of the color area is greater than a predetermined difference An image processing program for realizing a function of performing brightness correction on the image data with brightness lower than the target.

Images