WO2012056518A1 - 画像処理装置、画像処理方法及び画像処理プログラム - Google Patents
画像処理装置、画像処理方法及び画像処理プログラム Download PDFInfo
- Publication number
- WO2012056518A1 WO2012056518A1 PCT/JP2010/068975 JP2010068975W WO2012056518A1 WO 2012056518 A1 WO2012056518 A1 WO 2012056518A1 JP 2010068975 W JP2010068975 W JP 2010068975W WO 2012056518 A1 WO2012056518 A1 WO 2012056518A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- value
- luminance
- input
- conversion coefficient
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 86
- 238000003672 processing method Methods 0.000 title claims description 14
- 238000006243 chemical reaction Methods 0.000 claims abstract description 122
- 238000004364 calculation method Methods 0.000 claims abstract description 72
- 239000013598 vector Substances 0.000 claims abstract description 38
- 230000006870 function Effects 0.000 claims description 91
- 239000002131 composite material Substances 0.000 claims description 76
- 230000015572 biosynthetic process Effects 0.000 claims description 41
- 238000003786 synthesis reaction Methods 0.000 claims description 41
- 238000012937 correction Methods 0.000 claims description 39
- 238000012886 linear function Methods 0.000 claims description 19
- 238000000034 method Methods 0.000 description 79
- 238000010586 diagram Methods 0.000 description 26
- 230000002194 synthesizing effect Effects 0.000 description 25
- 239000000203 mixture Substances 0.000 description 18
- 238000011156 evaluation Methods 0.000 description 9
- 238000003384 imaging method Methods 0.000 description 7
- 238000013459 approach Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 102000051759 human factor J Human genes 0.000 description 2
- 108700008420 human factor J Proteins 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/387—Composing, repositioning or otherwise geometrically modifying originals
- H04N1/3876—Recombination of partial images to recreate the original image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/2624—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects for obtaining an image which is composed of whole input images, e.g. splitscreen
Definitions
- the present invention relates to an image processing apparatus, an image processing method, and an image processing program.
- Patent Documents 1 and 2 an apparatus that joins captured images to create a panoramic still image that is a single wide-angle still image is known (see, for example, Patent Documents 1 and 2).
- the small region is template-matched in the region where the first image and the second image overlap.
- Positioning is performed, gradation conversion is performed on the pixel value of at least one of the regions where the two images overlap, and the brightness in the region where the two images overlap is corrected so that the joint is not noticeable.
- the present invention has been made to solve such a technical problem, and an image that can perform processing to make the joint of the image inconspicuous at low load when stitching together images having different exposure conditions. It is an object to provide a processing device, an image processing method, and an image processing program.
- the image processing apparatus joins a first image configured by joining one image or a plurality of the images and an input second image each time the second image is input.
- An image processing apparatus that sequentially generates a composite image, wherein a motion vector is acquired based on the immediately preceding input image input immediately before the second image and the second image among the images constituting the first image,
- An overlapping area acquisition unit that acquires an overlapping area where the immediately preceding input image and the second image overlap based on the acquired motion vector, a luminance value of the first image in the overlapping area, and a luminance of the second image Based on the value, a luminance value conversion coefficient calculation that calculates a luminance conversion coefficient for converting the luminance value of the first image or the second image so that a change in the luminance value of the first image and the second image is small.
- a composite image generation unit that converts the first image or the second image using the luminance conversion coefficient, and generates the composite image by connecting the first image and the second image. It is prepared for.
- the overlapping region acquisition unit acquires a motion vector based on the second image and the immediately preceding input image input immediately before the second image, and based on the acquired motion vector, An overlapping area where the input image and the second image overlap is acquired, and the luminance value conversion coefficient calculation unit calculates the first image and the second image based on the luminance value of the first image and the luminance value of the second image in the overlapping area.
- a luminance conversion coefficient for converting the luminance value of the first image or the second image is calculated so that a change in the luminance value of the first image or the second image is small, and the synthesized image generation unit converts the first image or the second image based on the luminance conversion coefficient. After conversion, both images are stitched together.
- a luminance absolute value calculation unit that calculates a luminance absolute value of the second image, which is a ratio to the sum of luminance values, and the luminance value conversion coefficient calculation unit includes the first of the images constituting the first image.
- the luminance value conversion of the second image that depends on the distance from the joint between the target image and the second image based on the luminance absolute value of the target image overlapping the two images and the luminance absolute value of the second image It is preferable to calculate the coefficient.
- the luminance absolute value of the second image is one image constituting the first image, and the sum of the luminance values of the reference image that is used as a reference when calculating the luminance absolute value is used, without using the sum of the luminance values of the immediately preceding input image. Calculated based on the absolute luminance value. That is, it is not necessary to always record the sum of the luminance values of the reference image in order to calculate the absolute luminance value of the second image that is sequentially input. Therefore, the processing load can be reduced.
- the luminance value conversion coefficient calculation unit sets the luminance value conversion coefficient so that the weight of the second image at the synthesis position (predetermined position in the synthesis campus) increases in proportion to the distance from the joint of the synthesis position. It is preferable to calculate.
- the luminance value after conversion of the second image can be corrected so as to approach the original luminance value of the second image as the distance from the joint is increased.
- the composite image generation unit determines the luminance value conversion coefficient at the composite position based on the distance from the joint of the composite position, and calculates the brightness value of the second image based on the determined brightness value conversion coefficient.
- a correction function to be converted may be derived, and the luminance value of the second image used at the synthesis position may be calculated based on the correction function and the luminance value of the second image.
- the correction function is a linear function in which the luminance value conversion coefficient is a proportional constant in a range where the pixel value before conversion and the pixel value after conversion are less than a predetermined threshold, and the pixel value before conversion or In a range in which the pixel value after conversion is equal to or greater than a predetermined threshold, it is preferable that the linear function is continuous with the linear function up to the threshold and passes through the maximum pixel value.
- the correction function is a linear function that is continuous with the linear function up to the threshold and passes through the maximum pixel value. It is possible to generate a natural composite image while suppressing the calculation cost.
- a motion vector is acquired based on the image input immediately before the second image among the images constituting the first image, and the second image, and the second image is acquired based on the acquired motion vector.
- a center position acquisition unit that acquires position information of a first center point that is a center point of each image constituting one image and position information of a second center point that is a center point of the second image;
- the generation unit acquires the first center point of an image that overlaps the second image among images constituting the first image, and acquires the position information of the first center point and the position of the second center point. Based on the information, it is preferable to connect the obtained perpendicular bisector of the first center point and the second center point as a joint between the first image and the second image to generate the composite image. It is.
- the composite image generation unit is configured to obtain a pixel value within a range from a joint between the first image and the second image to a position separated by a predetermined distance and within a range narrower than the overlapping region. It is preferable to use a composite value of the pixel value of the first image and the pixel value of the second image.
- the pixel value from the joint to a position away from the joint by a predetermined distance is a composite value, so that the joint can be made inconspicuous.
- it is possible to calculate a composite value of pixel values using the second image after the luminance value is converted it is possible to avoid a significant change in the difference between the luminance values.
- the first region is reduced while reducing the processing load by narrowing the composition region near the joint. The difference between the brightness values of the first image and the second image that become noticeable by narrowing the region can be corrected based on the brightness values of the first image and the second image in a region wider than the region where the pixels are combined. .
- the image processing method joins a first image configured by joining one image or a plurality of the images and an input second image each time the second image is input.
- An overlapping area acquisition step for acquiring an overlapping area where the immediately preceding input image and the second image overlap based on the acquired motion vector, a luminance value of the first image in the overlapping area, and a luminance of the second image Based on the value, a luminance conversion coefficient calculation for calculating a luminance conversion coefficient for converting the luminance value of the first image or the second image so that a change in the luminance value of the first image and the second image is small.
- the image processing method according to the present invention has the same effects as those of the image processing apparatus of the present invention described above.
- the image processing program causes a computer to input a first image composed of one image or a plurality of the images and an input second image each time the second image is input.
- An image processing program that functions to sequentially generate a composite image by stitching together, based on the immediately preceding input image and the second image that are input immediately before the second image among the images constituting the first image
- An overlap area acquisition unit that acquires an overlap area where the immediately preceding input image and the second image overlap based on the acquired motion vector, a luminance value of the first image in the overlap area, and Based on the luminance value of the second image, the luminance value of the first image or the second image is converted so that the change in the luminance value of the first image and the second image is small.
- the image processing program according to the present invention has the same effects as those of the image processing apparatus of the present invention described above.
- the image processing apparatus is an apparatus that sequentially creates a single image by joining input images every time it is input. For example, a plurality of continuously captured images are joined in real time. This is suitably employed when generating a panoramic image having a wider angle than the captured image.
- the image processing apparatus according to the present embodiment is preferably mounted on a mobile terminal with limited resources such as a mobile phone, a digital camera, and a PDA (Personal Digital Assistant), but is not limited thereto. For example, it may be mounted on a normal computer system.
- a mobile terminal having a camera function will be described as an example of the image processing apparatus according to the present invention in consideration of ease of understanding.
- FIG. 1 is a functional block diagram of a mobile terminal 2 including an image processing apparatus 1 according to the present embodiment.
- a mobile terminal 2 shown in FIG. 1 is a mobile terminal carried by a user, for example, and has a hardware configuration shown in FIG.
- FIG. 2 is a hardware configuration diagram of the mobile terminal 2.
- the portable terminal 2 physically includes a main storage device such as a CPU (Central Processing Unit) 100, a ROM (Read Only Memory) 101, and a RAM (Random Access Memory) 102, a camera, a keyboard, and the like.
- the input device 103, the output device 104 such as a display, the auxiliary storage device 105 such as a hard disk, and the like are configured as a normal computer system.
- Each function of the portable terminal 2 and the image processing apparatus 1 to be described later causes the input device 103 and the output device 104 to be controlled under the control of the CPU 100 by reading predetermined computer software on hardware such as the CPU 100, the ROM 101, and the RAM 102. This is realized by operating and reading and writing data in the main storage device and the auxiliary storage device 105.
- the image processing apparatus 1 normally includes a CPU 100, a main storage device such as the ROM 101 and the RAM 102, an input device 103, an output device 104, an auxiliary storage device 105, and the like. It may be configured as a computer system.
- the mobile terminal 2 may include a communication module or the like.
- the mobile terminal 2 includes a camera 30, an image processing device 1, and a display unit 31.
- the camera 30 has a function of capturing an image.
- an image sensor or the like is used as the camera 30.
- the camera 30 has a continuous imaging function that repeatedly captures images at a predetermined interval from a timing specified by a user operation or the like, for example.
- the user can slide the camera 30 to capture continuous images that overlap at least vertically and horizontally.
- the camera 30 has a function of outputting a captured image to the image processing apparatus 1 every time it is captured.
- the image processing apparatus 1 includes an image input unit 10, a center position acquisition unit 11, a composite image generation unit 12, a center position storage unit 13, an overlapping region acquisition unit 14, a luminance absolute value calculation unit 15, and a luminance value conversion coefficient calculation unit 16. I have.
- the image input unit 10 has a function of inputting an image captured by the camera 30.
- the image input unit 10 has a function of inputting, for example, an image captured by the camera 30 every time it is captured.
- the image input unit 10 also has a function of saving the first input image in a first temporary storage area provided in the mobile terminal 2.
- the image input unit 10 has a function of saving images input continuously from the next time in a second temporary storage area provided in the mobile terminal.
- the second temporary storage area is updated each time a new image is input, and the first temporary storage area is overwritten and saved with an image (intermediate composite image) that is sequentially combined each time an image is input.
- the image stored in the first temporary storage area is described as a first image
- the image stored in the second temporary storage area is described as a second image.
- the overlapping area acquisition unit 14 has a function of detecting the motion (motion vector) of the camera based on the input image and the image (immediately input image) input immediately before the input image.
- the overlapping area acquisition unit 14 has a function of acquiring an overlapping area between the input image and the image input immediately before the input image based on the obtained motion vector.
- 3 and 4 are schematic diagrams for explaining the overlapping region. As shown in FIG. 3, an overlapping region R1 between the first image F1 and the second image F2 can be calculated based on the camera movement. Also, as shown in FIG. 4, when the first image is composed of a plurality of images F1 and F2, the image input immediately before the second image F3 and the second image F3 based on the movement of the camera.
- An overlapping region R2 with F2 can be calculated.
- the overlapping regions R1 and R2 are rectangular regions, and identification thereof is extremely easy.
- the overlapping area acquisition unit 14 has a function of outputting the overlapping area information to the luminance absolute value calculation unit 15.
- the luminance absolute value calculation unit 15 has a function of calculating the luminance absolute value of the input image.
- the luminance absolute value is a luminance value calculated based on the luminance value of a certain image.
- the reference image is not particularly limited, but here, the first image that is input first and stored in the first temporary storage area will be described as a reference.
- the luminance absolute value is calculated based on the sum of luminance values in the overlapping area specified by the overlapping area acquisition unit 14. In other words, the luminance absolute value calculation unit 15 calculates the sum of the luminance values of the second image included in the overlapping region and the luminance value of the immediately preceding input image included in the overlapping region, and the ratio of the sum of the luminance values of the two images.
- the absolute luminance value of the second image is calculated based on the absolute luminance value of the immediately preceding input image. Assuming that the sum of the luminance values of the second image included in the overlapping area is S 2 , the sum of the luminance values of the immediately preceding input images included in the overlapping area is S 1 , and the absolute luminance value of the immediately preceding input image is H 1 , the second image luminance absolute value of H 2 is calculated using equation 1 below. Note that a predetermined value is used as the absolute luminance value of the first image first input and stored in the first temporary storage area, for example, 1.
- the luminance absolute value calculation unit 15 has a function of outputting the calculated luminance absolute value to the luminance value conversion coefficient calculation unit 16.
- the center position acquisition unit 11 has a function of acquiring the position information of the center point of the image (the initial first image or the second image) input by the image input unit 10.
- the center point is a point that is uniquely determined from the outer edge of the image.
- the position information may be position information associated with the real space, or may be relative position information associated with images input continuously.
- the center position acquisition unit 11 has a function of detecting a camera motion (motion vector) based on the input image and the image input immediately before it in order to acquire the position information.
- the center position acquisition unit 11 has a function of calculating the position information of the center point of the input image based on the obtained motion vector and the position information of the center point of the image input immediately before. .
- the center position acquisition unit 11 acquires the position information of the center point only for the first input image (initial first image), and the subsequent input image (second image).
- the position information of the center point is acquired based on the motion vector obtained using the input image and the image immediately before the input. For example, for the second image input at the nth time (n> 1), a motion vector is acquired using the second image and the second image input at the (n-1) th time, and based on the acquired motion vector The position information of the center point of the second image input for the nth time is acquired.
- the center position acquisition unit 11 may calculate the motion vector by using the image which is not only the image input immediately before but the image input immediately before, and further reduces only the luminance element.
- the center point of each image constituting the first image will be described as the first center point
- the center point of the second image will be described as the second center point.
- the center position acquisition unit 11 has a function of outputting the acquired position information of the center point to the composite image generation unit 12.
- the composite image generation unit 12 has a function of generating a composite image obtained by connecting the input image (second image) and the already input image (first image). Part 122 is provided.
- the distance calculation unit 121 has a function of specifying an image that overlaps the second image among the images constituting the first image based on, for example, the motion vector acquired by the center position acquisition unit 11. Then, the distance calculation unit 121 specifies the first center point closest to the predetermined position of the image that overlaps the second image, and calculates the distance between the specified first center point and the second center point of the second image. It has a function.
- the predetermined position may be a position at a grid point arranged in a grid pattern. For example, lattice points are arranged in the composite image (here, the first image) to obtain the above-described predetermined position.
- the distance calculation unit 121 specifies the first center point closest to the grid point for each grid point before calculating the distance between the first center point and the second center point, and stores the first center point in the center position storage unit 13 in advance.
- Has the function of storing That is, the center position storage unit 13 stores a lattice point included in the first image and the first center point closest to the lattice point in association with each other.
- the distance calculation unit 121 identifies an image that overlaps the input second image among the images constituting the first image, refers to the center position storage unit 13, and is closest to the identified grid point. One center point is acquired.
- the distance calculation unit 121 calculates the distance between the first center point and the second center point that are different for each lattice point.
- the distance calculation unit 121 has a function of outputting the calculated distance to the synthesis unit 122.
- the distance calculation unit 121 also has a function of outputting the coordinate information and the like of the first center point and the second center point used for calculating the distance to the luminance value conversion coefficient calculation unit 16.
- the luminance value conversion coefficient calculation unit 16 has a function of calculating a luminance value conversion coefficient that is a coefficient of a correction function for converting a luminance value.
- the luminance value conversion coefficient calculation unit 16 joins the target image and the second image based on the luminance absolute value of the target image that overlaps the second image among the images constituting the first image and the luminance absolute value of the second image.
- the luminance value conversion coefficient calculation unit 16 specifies a vertical bisector that is a joint between the first image and the second image, based on positional information between the first center point and the second center point.
- the luminance value conversion coefficient calculation unit 16 calculates the luminance value conversion coefficient J using the following Expression 2.
- the brightness absolute value in H ex is combined position (predetermined position in the composite campus)
- H 2 is the brightness absolute value of the second image
- luminance absolute value of H f is the target image
- D 0 is the perpendicular bisector
- the luminance value conversion coefficient calculation unit 16 has a function of calculating the luminance value conversion coefficient so that the weight of the second image at the synthesis position increases in proportion to the distance from the joint at the synthesis position. .
- the luminance value conversion coefficient calculation unit 16 has a function of calculating the luminance value conversion coefficient so that the brightness value of a pixel range from the perpendicular bisector to a maximum combined position D 0 is converted.
- the luminance value conversion coefficient calculation unit 16 has a function of outputting the generated luminance value conversion coefficient to the synthesis unit 122.
- the synthesizing unit 122 has a function of joining the first image and the second image based on the distance between the first center point and the second center point calculated by the distance calculation unit 121. For example, the synthesizing unit 122 determines the pixel at the predetermined position based on the distance from the predetermined first position to the vertical bisector between the first center point and the second center point at the predetermined position in the combined image. Has a function to determine the value.
- 5 and 6 are schematic diagrams for explaining a perpendicular bisector between the first center point and the second center point. As shown in FIG. 5, a perpendicular bisector L1 can be drawn between the first center point P1 of the first image F1 and the second center point P2 of the second image F2.
- the distance to the vertical bisector L1 at a predetermined position in the composite image is calculated, and the pixel value at the predetermined position is determined.
- the first image is composed of a plurality of images F1 and F2
- each of the first center points P1 and P2 is bisected vertically with the second center point P3.
- Lines L2 and L3 can be drawn.
- a plurality of vertical bisectors can be drawn.
- a distance to a vertical bisector at a predetermined position in the composite image is calculated, and a pixel value at the predetermined position is determined.
- the composition unit 122 has a function of converting the pixel value of the second image before joining the second image to the first image. For example, the synthesis unit 122 determines the luminance value conversion coefficient J at the synthesis position based on the luminance value conversion coefficient J calculated by the luminance value conversion coefficient calculation unit 16 and the distance D from the joint to the synthesis position. . Then, the synthesis unit 122 derives a correction function Y that converts the luminance value of the second image based on the determined luminance value conversion coefficient J.
- the correction function Y is not particularly limited, and for example, a linear function is used.
- the synthesizing unit 122 calculates the luminance value of the second image used at the synthesis position based on the correction function Y and the luminance value of the second image.
- the synthesizing unit 122 executes the luminance value changing process of the second image before the image synthesizing process described later.
- the combining unit 122 uses the distance to the vertical bisector in order to adopt the pixel value of the image closest to the combining position as the pixel value of the combining position among the first image and the second image. .
- the synthesizing unit 122 uses the distance from a predetermined position in the synthesized image to the vertical bisector as an evaluation value for evaluating the proximity of the input image.
- the synthesizing unit 122 evaluates the proximity T of the input image using the following Expression 3.
- A is a distance from a predetermined position to be synthesized to the second center point
- B is a distance from the predetermined position to be synthesized to the nearest first center point
- C is a distance from the first center point to the second center point. The distance to the point.
- the synthesizing unit 122 determines the pixel value at the synthesis position (predetermined position) using the proximity T obtained by Expression 1 as an evaluation value. For example, when the distance from the predetermined position to the vertical bisector is greater than a predetermined value and the predetermined position is closer to the first center point than the second center point, the synthesizing unit 122 performs pixel processing for the first image. The value is the pixel value at the predetermined position. On the other hand, when the distance from the predetermined position to the vertical bisector is greater than the predetermined value and the predetermined position is closer to the second center point than the first center point, the synthesizing unit 122 performs pixel processing for the second image. The value is the pixel value at the predetermined position.
- the combining unit 122 combines the pixel value of the first image and the pixel value of the second image to generate a pixel at the predetermined position.
- Value As a method of combining, a conventional method can be employed. For example, a method of using an average value or a weighted average value of the pixel value of the first image and the pixel value of the second pixel as the pixel value at the predetermined position is used. It is done.
- the combining unit 122 determines whether the predetermined position of the combined image is closer to the first center point or the second center point with the vertical bisector as a boundary, and determines the first image and the second image. It has a function of determining which pixel value of the image is adopted. Then, when the predetermined position is in the vicinity of the vertical bisector, that is, for the predetermined position where the distance from the vertical bisector in the composite image is equal to or less than the predetermined value, the synthesizing unit 122 In addition, by combining the pixel values of the second image and the pixel values of the second image, the luminance difference between the joints is reduced to generate a composite image with less discomfort.
- the pixel value at a predetermined position is determined using the closeness T as an evaluation value.
- T the closeness
- the predetermined value W used to determine whether or not to combine pixel values.
- the pixel value of the first image and the pixel value of the second image are combined at an image position within 8 pixels with respect to the vertical bisector.
- the predetermined value W can be set to a range narrower than the overlapping region by setting it to about several pixels to several tens of pixels.
- the grid points in the composite image are arranged in a grid pattern so as to include the first image before the second image is input.
- the lattice points are newly added so as to include not only the first image but also the second image.
- the combining unit 122 can also read the first center point closest to the lattice point at high speed by referring to the center position storage unit 13.
- the synthesizing unit 122 has a function of determining the pixel value in the block surrounded by the lattice points based on the determination result at the lattice points in order to further increase the speed. For example, the synthesizing unit 122 is based on the distance from the lattice point arranged in a lattice pattern in the composite image to the perpendicular bisector between the first center point and the second center point closest to the lattice point. It has a function of determining pixel values at the lattice points.
- the synthesizing unit 122 adopts the pixel value of the first image for the block (region) surrounded by the grid points whose pixel values are the first image. That is, the synthesizing unit 122 does not perform any processing on the block and sets the next block as a processing target.
- the synthesizing unit 122 adopts the pixel value of the second image for the block surrounded by the grid points whose pixel values are the second image. That is, the composition unit 122 copies the second image as it is for the block.
- the pixel value is a composite value of the pixel values of the first image and the second image.
- the proximity T at the pixel position in the block is obtained by linear interpolation from the proximity T at the lattice point, and evaluated by the above-described evaluation method, thereby appropriately specifying the pixel position to be synthesized. Can do.
- the pixel value determination function of the combining unit 122 (which pixel value of the first image or the second image is used at a certain combining position, or the average of both images)
- the function of determining whether to use a value etc.) and the brightness value conversion function are described separately, and the example of determining the pixel value after converting the brightness value has been described.
- the brightness value is converted. May be.
- it may be configured such that the conversion of the luminance value and the determination of the pixel value are performed at the same time by using a single determination formula for exhibiting both functions.
- the distance D used in Equation 2 is 1 ⁇ 2 of the proximity T used in Equation 3.
- the distance D has an origin on the vertical bisector, takes a positive value when the synthesis position is on the first image side as viewed from the vertical bisector, and the synthesis position is second when viewed from the vertical bisector. When it is on the image side, it takes a negative value. For this reason, as shown below, it is possible to simultaneously determine the pixel value determination at a predetermined position and the presence / absence of luminance conversion using the distance D as an evaluation value.
- the predetermined value W is the same as the predetermined value W used in the above-described determination formula with the proximity T as an evaluation value, and W / 2 is larger than the maximum distance D 0 from the vertical bisector to the composite position. as smaller, predetermined value W and the maximum distance D 0 is preset. For example, 16 pixels as a predetermined value W (first threshold value) used for determining whether or not to combine, and an input image as a maximum distance D 0 (second threshold value) used for determining whether or not to perform luminance conversion 3/8 of the width is used.
- the synthesis unit 122 can increase the speed by first performing the above-described determination formula in units of blocks.
- the synthesizing unit 122 calculates a composition ratio (blend ratio) and a luminance conversion value for each pixel position, and thus the method of changing the pixel value and the luminance value is independent of each other. However, since the distance calculation and the pixel value determination process can be performed at the same time, it is possible to increase the speed by using the determination formula.
- the synthesizing unit 122 has a function of updating the nearest first center point recorded for each lattice point after generating a synthesized image by the above processing. For example, when a composite image is generated by connecting the first image and the second image, the closest first center point may be changed depending on the lattice points included in the composite image. For this reason, by performing the process of updating the nearest first center point after executing the joining process, accurate information on the nearest first center point can be maintained.
- the composition unit 122 overwrites and saves the composite image created by joining the first image and the second image in the first temporary storage area.
- the composition unit 122 stores the latest composite image in the first temporary storage area. That is, when there is a second image to be input next, a process of joining the second image to the latest composite image (intermediate composite image) is executed.
- the synthesizing unit 122 does not record and hold all the images to be synthesized, but refers to them by sequentially synthesizing the input images, so that the images can be synthesized with a small amount of memory.
- the synthesizing unit 122 has a function of outputting the synthesized image stored in the first temporary storage area to the display unit 31.
- the display unit 31 is connected to the composite image generation unit 12 and has a function of notifying the user of the output composite image. For example, a liquid crystal display or the like is used as the display unit 31.
- FIG. 7 is a flowchart showing the operation of the image processing apparatus 1 according to the present embodiment.
- the control process illustrated in FIG. 7 is executed, for example, at the timing when the imaging function of the mobile terminal 2 is turned on, and is repeatedly executed at a predetermined cycle.
- FIGS. 8 and 13 are schematic diagrams when one image is connected to one already input image
- FIG. 9 is a schematic diagram explaining the luminance absolute value
- FIG. 10 is a diagram explaining the luminance value conversion coefficient.
- FIGS. 11 and 12 are schematic diagrams for explaining a correction function
- FIGS. 14 and 15 are schematic diagrams when one image is joined to an image that has already been input and synthesized.
- the image processing apparatus 1 executes an initial process (S12).
- the image input unit 10 inputs an image F1 from the camera 30 and stores the image F1 in the first temporary storage area as the first image F1.
- the center position acquisition unit 11 acquires position information of the first center point P1, which is the center point of the first image F1.
- the synthesized image generation unit 12 arranges grid points K n (n: integer) in a grid pattern on the synthesized campus (synthesized image) so as to include the first image F1.
- the composite image generating unit 12 specifies the first center point P1 as a first center point closest to the grid point K n, each grid point K n in association a first center point P1 center position storage section 13 To record. This completes the initial process.
- the process of S12 proceeds to the second image input process (S14).
- the image input unit 10 inputs the image F2 from the camera 30, and stores it as the second image F2 in the second temporary storage area.
- the second image F2 is an image of the same size captured at an imaging position different from the imaging position of the first image F1, and has an overlapping area with the first image F1.
- the process of S14 proceeds to the acquisition process of the center point position and the overlapping area acquisition process (S15, S17).
- the center point position acquisition process and the overlap area acquisition process are processed in parallel thereafter. First, the process of acquiring the position of the center point will be described.
- the center position acquisition unit 11 acquires the position information of the second center point P2, which is the center point of the second image F2. For example, the center position acquisition unit 11 acquires the position information of the second center point P2 based on the motion vectors of the first image F1 and the second image F2.
- the process of S15 ends, the process proceeds to a distance acquisition process between the center points (S16).
- the distance calculation unit 121 determines the first center based on the position information of the first center point P1 obtained in the process of S12 and the position information of the second center point P2 obtained in the process of S16. A distance between the point P1 and the second center point P2 is calculated. As shown in FIG. 8, the distance C between the first center point P1 of the first image F1 and the second center point P2 of the second image F2 is calculated. By calculating the distance C between the center points, the distance between the first center point P1 and the second center point P2 to the vertical bisector L1 is evaluated, and the vertical bisector L1 is determined as the image F1 and the image F1. It can be connected to F2. When the process of S16 ends, the process proceeds to a luminance value conversion coefficient calculation process (S19).
- the overlapping area acquisition unit 14 acquires the overlapping area of the first image input in the process of S12 and the second image input in the process of S14.
- the overlapping region acquisition unit 14 acquires the overlapping region R1 based on the motion vectors of the first image F1 and the second image F2.
- the process proceeds to a luminance absolute value calculation process (S18).
- the luminance absolute value calculation unit 15 calculates the luminance absolute value H based on the overlapping region R1 acquired in the process of S17.
- FIG. 9 is a graph showing the luminance value at the combined position of each image, where the horizontal axis is the combined campus position and the vertical axis is the luminance value. As shown in FIG. 9, it is assumed that the first image F1 and the second image F2 overlap in the overlapping region R1. Here, with respect to the first image F1 which is initially input to the luminance absolute value H 1 of the first image F1 and 1.
- the ratio (S 2 / S 1 ) of the sum S 1 of pixel values in the overlapping region R 1 of the first image F 1 to the sum S 2 of pixel values in the overlapping region R 1 of the second image F 2 is 1.2 here.
- the brightness absolute value calculating unit 15 using Equation 1 described above, the brightness absolute value of H 2 second image F2 is calculated as 1.2.
- the luminance value conversion coefficient calculation unit 16 specifies the vertical bisector of the first center point and the second center point from the position information of the center point acquired in the processes of S15 and S16. Then, the luminance value conversion coefficient calculation unit 16 calculates the absolute luminance value H 2 of the second image F2 specified in the processing of S18 and the absolute luminance value H f of the first image F1 stored in the absolute luminance value storage unit.
- the luminance value conversion coefficient J is calculated based on For example, the luminance value conversion coefficient calculation unit 16, before calculating the luminance value conversion factor J, keep the brightness absolute value H f of the first image F1 and stored in the luminance absolute value storage unit.
- the luminance value conversion coefficient calculation unit 16 combines the absolute luminance value H f after the composition by multiplying the absolute luminance value H 2 of the second image F2 by the luminance value conversion coefficient J into the absolute luminance value storage unit after the synthesis. Is stored in the luminance absolute value storage unit of the first image F1.
- FIG. 10 is a schematic diagram illustrating the determination process of the luminance value conversion coefficient J of the second image.
- the absolute luminance value H f of an image constituting the first image F1 is 1.1
- the absolute luminance value H 2 of the second image F2 input next to this image is 1.2. To do.
- the luminance value conversion coefficient calculation unit 16 uses the equation 2 to calculate the luminance for the composite position G1 existing on the vertical bisector L1 because the distance D from the vertical bisector L is 0.
- the value conversion coefficient J is calculated as 0.92.
- the synthetic position G2 when the weighting D / D 0 of the input image and 0.7, calculated as 0.97 the luminance value conversion factor J using Equation 2.
- the weight D / D 0 of the input image is set to 1, and the luminance value conversion coefficient J is calculated to be 1.00 using Equation 2.
- the composition unit 122 corrects the second image by setting the correction function Y, and then combines the images F1 and F2 to generate a composite image.
- the setting of the correction function Y will be described.
- 11 and 12 are schematic diagrams for explaining the correction functions Y1 and Y2, where the horizontal axis indicates the input image value and the vertical axis indicates the final pixel value. That is, the correction functions Y1 and Y2 are functions that return the value of the final pixel value when an input pixel value is given.
- the synthesizer 122 sets the correction functions Y1 and Y2 with the luminance value conversion coefficient J as the slope of the correction functions Y1 and Y2, that is, the luminance value conversion coefficient J as a proportional constant of a linear function.
- the correction function is a linear function
- the difference between the input pixel value and the final pixel value increases as the input pixel value increases regardless of the magnitude of the luminance value conversion coefficient J.
- the correction function Y1 converts the final pixel values to “58” and “253”, respectively, and the differences are “8” and “33”.
- the correction function Y2 the final pixel values are converted to “40” and “176”, and the differences are “10” and “44”, respectively.
- the final pixel value is constant at 255 for an input pixel value greater than a certain value, and there is a possibility that the information of the original pixel value may not be appropriately expressed. Therefore, for example, as illustrated in FIG. 12, the combining unit 122 finally passes the maximum pixel value (255) in a range where the pixel value before conversion or the pixel value after conversion is equal to or greater than a predetermined threshold value. As described above, the correction functions Y1 and Y2 are bent at the intersection of the threshold values.
- the range from 0 to the threshold value is a linear function having the luminance value conversion coefficient J as a slope
- the threshold value to the final pixel value is a linear function that is continuous with the linear function up to the threshold value and passes through the maximum pixel value.
- the composition unit 122 inputs the second image after setting the correction function Y, and converts the pixel value.
- the pixel value here is expressed in, for example, a YUV space, and the pixel value is also changed by changing the luminance value.
- the synthesis unit 122 joins the image F1 and the converted image F2 to generate a synthesized image.
- the first image F1 and the second image F2 are arranged in the coordinate space of the composite image. And about the area
- the additional grid points K m, combining unit 122 identifies the first center point P1 as a first center point closest to the grid point K m, associating a first center point P1 to the respective grid point K m And recorded in the center position storage unit 13. At this time, a point at infinity is set in P1.
- the synthesis unit 122 evaluates the distance from the vertical bisector L1 for each grid point K n , K m arranged in the first image F1 and the second image F2, and the grid point K n , determining the pixel values in the K m. For example, to calculate a closeness T using equation 3 described above in order from the lattice point K n located at the upper left. Then, as a 16 predetermined value W which is used to determine whether the synthesis of the pixel value, to assess the closeness T, determining the pixel values at the grid points K n.
- the synthesis unit 122 refers to the center position storage unit 13 and acquires the closest first center point P1, and the distance C between the first center point P1 and the second center point P2. Then, the distance A to the first center point P1 and the distance B to the second center point P2 are calculated, and the proximity T is calculated using Equation 3. Similar processing is performed for the lattice point X2. Combining unit 122, for the grid point X1, X2, since a grid point K n which are arranged in the first image F1, performs a process of evaluation by calculating a closeness T as described above. On the other hand, for the newly added lattice point K m , the proximity T is clearly below the threshold value, so the proximity T of these lattice points K m is set to ⁇ , and the calculation of the proximity T is omitted.
- the synthesizing unit 122 evaluates the calculated proximity T for each grid point K n , and all the proximity T of each of the grid points K n are larger than 16 for the blocks surrounded by the grid points K n at the four corners. In this case, the processing is skipped for the block. For example, for the lattice points X3 to X6, since the closeness T is all greater than 16, processing is skipped for blocks surrounded by the lattice points X3 to X6. On the other hand, 4 closeness T corners of the grid points K n is when all smaller than -16 employs the pixel value of the second image F2 as the pixel value of the block.
- the second image F2 is copied for the block surrounded by the lattice points X7 to X10. Then, 4 if closeness T corners of the grid points K n is not greater than all 16, or 4 if closeness T corners of the grid points K n is not less than all -16, pixel of the block
- the pixel values of the first image F1 and the second image F2 are synthesized as values. For example, the proximity T is larger than 0 for the lattice points X11 and X12, and the proximity T is smaller than 0 for the lattice points X13 and X14.
- the pixel values of the first image F1 and the second image F2 are synthesized.
- the proximity T at the pixel position in the block is linearly interpolated with the proximity T of X11 to X14, the proximity T is calculated at each pixel position, and the threshold W is evaluated.
- This evaluation method is the same as described above. Then, for the pixel position of closeness T that is greater than or equal to the threshold value ⁇ W and less than or equal to the threshold value W, the average of the pixel value of the first image F1 and the pixel value of the second image F2 is calculated, and the pixel value at the pixel position is calculated.
- the synthesizing unit 122 identifies the pixel value at the lattice point using a determination formula using the distance D instead of the proximity T. For example, to calculate a closeness T using equation 3 above from the lattice point K n located at the upper left in order to calculate the distance D from the closeness T.
- a predetermined value W used for determining whether or not to synthesize pixel values is set to 16
- a maximum distance D 0 used for determining whether or not to perform luminance conversion is set to 240 (input image width of 640 pixels 3 / 8), and to evaluate the distance D, and determining the pixel values at the grid points K n.
- the synthesis unit 122 refers to the center position storage unit 13 and acquires the closest first center point P1, and the distance C between the first center point P1 and the second center point P2.
- the distance A to the first center point P1 and the distance B to the second center point P2 are calculated
- the proximity T is calculated using Equation 3
- the distance D is calculated from the proximity T. Similar processing is performed for the lattice point X2.
- the synthesizing unit 122 evaluates the calculated distance D for each grid point K n , and when all the distances D of the grid points K n are larger than 8 for the blocks surrounded by the grid points K n at the four corners ( That is, when the absolute value of the distance D is larger than 8 and the combined position is closer to the first center point than the second center point), the processing is skipped for the block.
- the distances D are all greater than 8 for the lattice points X3 to X6. Therefore, for the block surrounded by the lattice points X3 to X6, the pixels of the first image are adopted, and the processing is skipped.
- the distance D of the corners of the grid points K n is less than all -240 (i.e., the absolute value of the distance D is greater than 240, and the synthesis position in the second center point than the first center point If it is close, the pixel value of the second image F2 is copied as the pixel value of the block.
- the four corners of the grid points K n distance D are all -240 or more and smaller than 8 (i.e., the absolute value of the distance D is larger than 8, and at 240 or less, and the combination position When closer to the second center point than the first center point), the pixel value of the second image F2 that has been subjected to luminance conversion by the correction function Y is copied as the pixel value of the block.
- the four corners of the distance D of grid points K n all -8 or more and, in the case of 8 or less (i.e., when the absolute value of the distance D is 8 or less), as the pixel value of the block, the The pixel value of one image and the pixel value of the second image F2 subjected to luminance conversion by the correction function Y are synthesized (weighted average). Then, 4 when the corners of the determination of the grid points K n is not all the same, the distance D at a pixel position in the block, and linear interpolation distance D 4 corners of the grid points K n, at each pixel location distance to calculate the D, and evaluated by the predetermined value W and the maximum distance D 0. This evaluation method is the same as described above.
- the combining unit 122 is a process of updating the first center point P ⁇ b> 1 recorded in the center position storage unit 13. Since the image F2 is synthesized, there are two first center points P1 and P2. Thus, combining unit 122, the first image to update the first center point closest to the grid point K n stored in the first temporary storage area. For example, in the case of the grid point X2, since the first center point P1 is closer to the first center point P1, P2, the update is not executed because it is the same as the previous time. On the other hand, for example, in the case of the grid point X1, since the first center point P2 is closer to the first center points P1 and P2, the stored information in the center position storage unit 13 is updated.
- the process of S22 ends, the process proceeds to a process for determining the presence or absence of an input image (S24).
- the image input unit 10 determines whether there is an image to be further input. For example, when the current imaging count is smaller than the auto continuous imaging count, it is determined that an input image exists. If it is determined in step S24 that an input image exists, the process proceeds to image input processing again (S14). Then, for example, the image F3 is input and stored in the second temporary storage area. Then, the center position acquisition unit 11 acquires the position of the center point P3 of the image F3 (S15). The distance calculation unit 121 then determines the distance between the first center points P1 and P2 of the composite image composed of the images F1 and F2 stored in the first temporary storage area and the second center point P3 of the input image F2. Are respectively calculated (S16).
- the overlapping area acquisition unit 14 acquires an overlapping area R2 between the image F2 and the image F3 (S17).
- the luminance absolute value calculation unit 15 calculates the luminance absolute value of the image F3 using Equation 1 (S18).
- the luminance absolute value is based on the first input image, but the information on the reference image is not used for the calculation of the luminance absolute value.
- the luminance value conversion coefficient calculation unit 16 of the image F3 calculates the luminance value conversion coefficient of the image F3 (S19).
- the composition unit 122 converts the image F3 with the correction function
- the images F1 and F2 and the image F3 are connected to generate a composite image.
- the first images F1, F2 and the second image F3 are arranged in the coordinate space of the composite image.
- new grid point K m (m: integer, dotted line in the figure) is added The Then, the additional grid points K m, combining unit 122, identifies the first central point P3 as the first center point closest to the grid point K m, associating a first central point P3 to each grid point K m And recorded in the center position storage unit 13.
- the synthesizing unit 122 has lattice points K n and K m (for example, X15, X16, and the like) arranged in the first image F1 and F2 and the second image F3.
- the distance from the vertical bisectors L2 and L3 is evaluated every X17, etc., and the pixel values at the lattice points K n and K m are specified.
- an image in which the images F1, F2, and F3 are combined is generated.
- the synthesis unit 122 updates the center point position of the grid point K n (S22). As described above, when there is an input image, the processes of S14 to S24 are repeatedly executed.
- step S24 if it is determined in step S24 that there is no input image, the process proceeds to display processing (S26).
- the image processing apparatus 1 outputs the composite image stored in the first temporary storage area to the display unit 31 for display.
- the image processing apparatus 1 may cut out both ends of the composite image, adjust the size, and output the combined image to the display unit 31.
- the control process shown in FIG. 7 ends.
- the process of S26 may be performed every time one image is input (that is, between S20 and S24).
- the image processing program includes a main module, an input module, and an arithmetic processing module.
- the main module is a part that comprehensively controls image processing.
- the input module operates the mobile terminal 2 so as to acquire an input image.
- the arithmetic processing module includes a center position acquisition module, a distance calculation module, an overlapping area acquisition module, a luminance absolute value calculation module, a luminance value conversion coefficient calculation module, and a synthesis module.
- the functions realized by executing the main module, the input module, and the arithmetic processing module are the image input unit 10, the center position acquisition unit 11, the distance calculation unit 121, the synthesis unit 122, and the overlap region acquisition of the image processing apparatus 1 described above.
- the functions of the unit 14, the luminance absolute value calculation unit 15, and the luminance value conversion coefficient calculation unit 16 are the same.
- the image processing program is provided by a storage medium such as a ROM or a semiconductor memory, for example.
- the image processing program may be provided as a data signal via a network.
- the overlapping area acquisition unit 14 moves based on the second image and the immediately preceding input image input immediately before the second image.
- a vector is acquired, and based on the acquired motion vector, an overlapping region where the immediately preceding input image and the second image overlap is acquired, and the luminance value conversion coefficient calculation unit 16 acquires the luminance value of the first image in the overlapping region and the first image.
- a luminance value conversion coefficient J for converting the luminance value of the second image so as to reduce the change in the luminance values of the first image and the second image is calculated.
- the first image and the second image converted based on the luminance value conversion coefficient J are joined together.
- the absolute luminance value is calculated based on the sum of the luminance values
- the luminance conversion coefficient is calculated using the absolute luminance value. Therefore, a robust calculation can be performed with a simple configuration.
- the luminance absolute value of the second image is one image constituting the first image, and the sum of the luminance values of the reference image that is used as a reference when calculating the luminance absolute value is used, without using the sum of the luminance values of the immediately preceding input image. Calculated based on the absolute luminance value. That is, it is not necessary to always record the sum of the luminance values of the reference image in order to calculate the absolute luminance value of the second image that is sequentially input. Therefore, the processing load can be reduced. Furthermore, when the luminance value conversion coefficient is calculated, the difference can be easily managed by using the luminance absolute value.
- the brightness value after the conversion of the second image is increased as the distance from the vertical bisector L that is a joint increases. Correction can be made so as to approach the original luminance value.
- the calculation cost of the correction function Y can be suppressed by adopting a linear function as the correction function Y.
- the correction function Y of the linear function is employed, the difference between the pixel value before conversion and the pixel value after conversion increases as the pixel value before conversion increases, and as a result, an unnatural composite image may be generated. is there. For this reason, in a range where the pixel value before conversion or the pixel value after conversion is equal to or greater than a predetermined threshold value, the correction function Y is corrected to a linear function that is continuous with the linear function up to the threshold value and passes through the maximum pixel value. A natural composite image can be generated while reducing the calculation cost of the function Y.
- the image processing device 1 when the first image and the input second image are sequentially combined, the first center point and the second center point Since the vertical bisector L can be used as a joint between images, it is possible to realize the sequential synthesis process at high speed and with a low load.
- the pixel value from the joint to a position that is a predetermined distance away is the composite value, so that the joint is not conspicuous. Can do.
- it is possible to calculate a composite value of pixel values using the second image after the luminance value is converted it is possible to avoid a significant change in the difference between the luminance values.
- the first region is reduced while reducing the processing load by narrowing the composition region near the joint.
- the difference between the brightness values of the first image and the second image that become noticeable by narrowing the area can be corrected based on the brightness values of the first image and the second image in an area wider than the area where the pixels are combined. .
- the above-described embodiment shows an example of the image processing apparatus according to the present invention.
- the image processing apparatus according to the present invention is not limited to the image processing apparatus 1 according to the embodiment, and the image processing apparatus according to the embodiment may be modified or otherwise changed without changing the gist described in each claim. It may be applied to the above.
- the camera 30 may capture a moving image.
- the image input unit 10 may have a function of extracting continuous images from the captured moving image.
- the image input by the image input unit 10 may be an image transmitted from another device via a network.
- the size of the image captured by the camera 30 is described as being the same. However, the size of the captured image may be different for each imaging.
- the area surrounded by the lattice points is described as a rectangle, but it may be a triangle or another polygon.
- the center position acquisition unit 11 calculates the motion vector using the input image and the image input immediately before has been described.
- the motion vector calculation method is not limited thereto. Is not something
- the motion vector may be calculated using the input image and the synthesized image generated so far.
- the example in which the luminance value of the second image is operated when the first image and the second image overlap each other has been described.
- the luminance value of the first image is operated, or It may be a case where the brightness values of both the first image and the second image are manipulated.
- SYMBOLS 1 DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 10 ... Image input part, 11 ... Center position acquisition part, 12 ... Composite image generation part, 121 ... Distance calculation part, 122 ... Composition part, 13 ... Center position memory
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Studio Devices (AREA)
- Studio Circuits (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
Description
Claims (9)
- 1枚の画像又は前記画像を複数つなぎ合わせて構成される第1画像と入力された第2画像とを当該第2画像の入力の度につなぎ合わせて合成画像を逐次生成する画像処理装置であって、
前記第1画像を構成する画像のうち前記第2画像の直前に入力された直前入力画像及び前記第2画像に基づいて動きベクトルを取得し、取得された前記動きベクトルに基づいて前記直前入力画像と前記第2画像とが重なる重なり領域を取得する重なり領域取得部と、
前記重なり領域における前記第1画像の輝度値及び前記第2画像の輝度値に基づいて、前記第1画像及び前記第2画像の輝度値の変化が小さくなるように前記第1画像又は前記第2画像の輝度値を変換する輝度変換係数を算出する輝度値変換係数算出部と、
前記輝度変換係数を用いて前記第1画像又は前記第2画像を変換し、前記第1画像と前記第2画像とをつなぎ合わせて前記合成画像を生成する合成画像生成部と、
を備えることを特徴とする画像処理装置。 - 前記重なり領域における前記直前入力画像の輝度値の総和及び前記第2画像の輝度値の総和に基づいて、前記第1画像のうち所定の基準画像の輝度値の総和と前記第2画像の輝度値の総和との比である前記第2画像の輝度絶対値を算出する輝度絶対値算出部を備え、
前記輝度値変換係数算出部は、前記第1画像を構成する画像のうち前記第2画像と重なる対象画像の前記輝度絶対値及び前記第2画像の前記輝度絶対値に基づいて、前記対象画像と前記第2画像とのつなぎ目からの距離に依存する前記第2画像の輝度値変換係数を算出する請求項1に記載の画像処理装置。 - 前記輝度値変換係数算出部は、合成位置における前記第2画像の重みが合成位置のつなぎ目からの距離に比例して大きくなるように前記輝度値変換係数を算出する請求項1又は2に記載の画像処理装置。
- 合成画像生成部は、合成位置のつなぎ目からの距離に基づいて合成位置での前記輝度値変換係数を決定し、決定された前記輝度値変換係数に基づいて前記第2画像の輝度値を変換する補正関数を導出し、前記補正関数及び前記第2画像の輝度値に基づいて合成位置で用いる前記第2画像の輝度値を算出する請求項1~3の何れか一項に記載の画像処理装置。
- 前記補正関数は、変換前の画素値及び変換後の画素値が所定の閾値未満となる範囲においては、前記輝度値変換係数を比例定数とする一次関数であり、変換前の画素値又は変換後の画素値が所定の閾値以上となる範囲においては、閾値までの一次関数と連続し最大画素値を通る一次関数である請求項4に記載の画像処理装置。
- 前記第1画像を構成する画像のうち前記第2画像の直前に入力された画像、及び前記第2画像に基づいて動きベクトルを取得し、取得された前記動きベクトルに基づいて、前記第1画像を構成する画像それぞれの中心点である第1中心点の位置情報、及び前記第2画像の中心点である第2中心点の位置情報を取得する中心位置取得部を備え、
前記合成画像生成部は、前記第1画像を構成する画像のうち前記第2画像と重なる画像の前記第1中心点を取得し、取得された前記第1中心点の位置情報及び前記第2中心点の位置情報に基づいて、取得された前記第1中心点と前記第2中心点との垂直二等分線を前記第1画像及び前記第2画像のつなぎ目としてつなぎ合わせ前記合成画像を生成する請求項1~5の何れか一項に記載の画像処理装置。 - 前記合成画像生成部は、前記第1画像と前記第2画像とのつなぎ目から所定距離離れた位置までの範囲内であって、前記重なり領域よりも狭い範囲内にある画素値を、前記第1画像の画素値及び前記第2画像の画素値の合成値とする請求項6に記載の画像処理装置。
- 1枚の画像又は前記画像を複数つなぎ合わせて構成される第1画像と入力された第2画像とを当該第2画像の入力の度につなぎ合わせて合成画像を逐次生成する画像処理方法であって、
前記第1画像を構成する画像のうち前記第2画像の直前に入力された直前入力画像及び前記第2画像に基づいて動きベクトルを取得し、取得された前記動きベクトルに基づいて前記直前入力画像と前記第2画像とが重なる重なり領域を取得する重なり領域取得ステップと、
前記重なり領域における前記第1画像の輝度値及び前記第2画像の輝度値に基づいて、前記第1画像及び前記第2画像の輝度値の変化が小さくなるように前記第1画像又は前記第2画像の輝度値を変換する輝度変換係数を算出する輝度変換係数算出ステップと、
前記輝度変換係数を用いて前記第1画像又は前記第2画像を変換し、前記第1画像と前記第2画像とをつなぎ合わせて前記合成画像を生成する合成画像生成ステップと、
を備えることを特徴とする画像処理方法。 - コンピュータを、1枚の画像又は前記画像を複数つなぎ合わせて構成される第1画像と入力された第2画像とを当該第2画像の入力の度につなぎ合わせて合成画像を逐次生成するように機能させる画像処理プログラムであって、
前記第1画像を構成する画像のうち前記第2画像の直前に入力された直前入力画像及び前記第2画像に基づいて動きベクトルを取得し、取得された前記動きベクトルに基づいて前記直前入力画像と前記第2画像とが重なる重なり領域を取得する重なり領域取得部、
前記重なり領域における前記第1画像の輝度値及び前記第2画像の輝度値に基づいて、前記第1画像及び前記第2画像の輝度値の変化が小さくなるように前記第1画像又は前記第2画像の輝度値を変換する輝度変換係数を算出する輝度値変換係数算出部、及び
前記輝度変換係数を用いて前記第1画像又は前記第2画像を変換し、前記第1画像と前記第2画像とをつなぎ合わせて前記合成画像を生成する合成画像生成部
として機能させることを特徴とする画像処理プログラム。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2010/068975 WO2012056518A1 (ja) | 2010-10-26 | 2010-10-26 | 画像処理装置、画像処理方法及び画像処理プログラム |
US13/260,278 US8538192B2 (en) | 2010-10-26 | 2010-10-26 | Image processing device, image processing method and storage medium |
JP2010542863A JP4941950B1 (ja) | 2010-10-26 | 2010-10-26 | 画像処理装置、画像処理方法及び画像処理プログラム |
EP10848135.9A EP2613290B1 (en) | 2010-10-26 | 2010-10-26 | Image processing device, image processing method, and image processing program |
CN201080014815.5A CN102741878B (zh) | 2010-10-26 | 2010-10-26 | 图像处理装置、图像处理方法以及图像处理程序 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2010/068975 WO2012056518A1 (ja) | 2010-10-26 | 2010-10-26 | 画像処理装置、画像処理方法及び画像処理プログラム |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012056518A1 true WO2012056518A1 (ja) | 2012-05-03 |
Family
ID=45993274
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2010/068975 WO2012056518A1 (ja) | 2010-10-26 | 2010-10-26 | 画像処理装置、画像処理方法及び画像処理プログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US8538192B2 (ja) |
EP (1) | EP2613290B1 (ja) |
JP (1) | JP4941950B1 (ja) |
CN (1) | CN102741878B (ja) |
WO (1) | WO2012056518A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140016866A1 (en) * | 2012-07-10 | 2014-01-16 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7983835B2 (en) | 2004-11-03 | 2011-07-19 | Lagassey Paul J | Modular intelligent transportation system |
US20090310149A1 (en) * | 2008-06-13 | 2009-12-17 | Kyocera Mita Corporation | Image forming apparatus |
SG190962A1 (en) * | 2010-12-29 | 2013-07-31 | Orthodyne Electronics Corp | Methods and systems for aligning tooling elements of ultrasonic bonding systems |
US8902313B2 (en) * | 2011-11-02 | 2014-12-02 | Robert Bosch Gmbh | Automatic image equalization for surround-view video camera systems |
US20130169663A1 (en) * | 2011-12-30 | 2013-07-04 | Samsung Electronics Co., Ltd. | Apparatus and method for displaying images and apparatus and method for processing images |
JP2013196454A (ja) * | 2012-03-21 | 2013-09-30 | Honda Elesys Co Ltd | 画像処理装置、画像処理方法および画像処理プログラム |
JP5929849B2 (ja) * | 2013-07-08 | 2016-06-08 | 株式会社デンソー | 画像生成装置、画像生成方法及び画像生成プログラム |
CN104125410B (zh) * | 2014-07-02 | 2017-04-12 | 深圳英飞拓科技股份有限公司 | 一种全景视频多镜头曝光补偿方法及其装置 |
JP6536183B2 (ja) * | 2015-06-01 | 2019-07-03 | 富士ゼロックス株式会社 | 画像読取装置及びプログラム |
JP6551316B2 (ja) * | 2016-06-10 | 2019-07-31 | 京セラドキュメントソリューションズ株式会社 | 画像読取装置および画像形成装置 |
JP6932934B2 (ja) * | 2017-01-23 | 2021-09-08 | セイコーエプソン株式会社 | スキャナー、スキャンプログラムおよびスキャンデータの生産方法 |
US11049218B2 (en) * | 2017-08-11 | 2021-06-29 | Samsung Electronics Company, Ltd. | Seamless image stitching |
CN109087606A (zh) * | 2018-08-15 | 2018-12-25 | 苏州佳世达光电有限公司 | 大尺寸显示屏幕的均匀度调整方法及系统 |
CN113256516A (zh) * | 2021-05-18 | 2021-08-13 | 中国科学院长春光学精密机械与物理研究所 | 一种图像增强方法 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09321972A (ja) | 1996-05-28 | 1997-12-12 | Canon Inc | 画像合成装置及び方法 |
JPH1091765A (ja) | 1996-09-10 | 1998-04-10 | Canon Inc | 画像合成装置及びその方法 |
JPH11102430A (ja) * | 1997-09-25 | 1999-04-13 | Sharp Corp | 画像処理装置 |
JPH11196319A (ja) * | 1997-12-26 | 1999-07-21 | Minolta Co Ltd | 撮像装置 |
JP2002042109A (ja) * | 2000-07-21 | 2002-02-08 | Topcon Corp | 医用画像合成処理装置及び方法並びに記憶媒体 |
JP2006119730A (ja) * | 2004-10-19 | 2006-05-11 | Seiko Epson Corp | 画像のつなぎ合わせ |
JP2006345400A (ja) * | 2005-06-10 | 2006-12-21 | Matsushita Electric Ind Co Ltd | ビデオカメラ装置 |
JP2009033224A (ja) * | 2007-07-24 | 2009-02-12 | Nippon Hoso Kyokai <Nhk> | 合成画像生成装置および合成画像生成プログラム |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6205259B1 (en) * | 1992-04-09 | 2001-03-20 | Olympus Optical Co., Ltd. | Image processing apparatus |
DE69416717T2 (de) * | 1993-05-21 | 1999-10-07 | Nippon Telegraph And Telephone Corp., Tokio/Tokyo | Bewegtbildkoder und -dekoder |
US5982951A (en) | 1996-05-28 | 1999-11-09 | Canon Kabushiki Kaisha | Apparatus and method for combining a plurality of images |
JP3851384B2 (ja) * | 1996-09-18 | 2006-11-29 | シャープ株式会社 | 画像合成装置および方法 |
US6215914B1 (en) | 1997-06-24 | 2001-04-10 | Sharp Kabushiki Kaisha | Picture processing apparatus |
US6720997B1 (en) | 1997-12-26 | 2004-04-13 | Minolta Co., Ltd. | Image generating apparatus |
JP2004334843A (ja) * | 2003-04-15 | 2004-11-25 | Seiko Epson Corp | 複数の画像から画像を合成する方法 |
US7260258B2 (en) * | 2003-06-12 | 2007-08-21 | Fuji Xerox Co., Ltd. | Methods for multisource color normalization |
JP4377932B2 (ja) * | 2007-07-26 | 2009-12-02 | 株式会社モルフォ | パノラマ画像生成装置およびプログラム |
US7916177B2 (en) * | 2007-08-03 | 2011-03-29 | Panasonic Corporation | Image-capturing apparatus, image-capturing method and program for detecting and correcting image blur |
JP5179223B2 (ja) * | 2008-02-26 | 2013-04-10 | オリンパス株式会社 | 撮像装置及び撮像プログラム |
-
2010
- 2010-10-26 CN CN201080014815.5A patent/CN102741878B/zh not_active Expired - Fee Related
- 2010-10-26 JP JP2010542863A patent/JP4941950B1/ja active Active
- 2010-10-26 US US13/260,278 patent/US8538192B2/en active Active
- 2010-10-26 WO PCT/JP2010/068975 patent/WO2012056518A1/ja active Application Filing
- 2010-10-26 EP EP10848135.9A patent/EP2613290B1/en not_active Not-in-force
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH09321972A (ja) | 1996-05-28 | 1997-12-12 | Canon Inc | 画像合成装置及び方法 |
JPH1091765A (ja) | 1996-09-10 | 1998-04-10 | Canon Inc | 画像合成装置及びその方法 |
JPH11102430A (ja) * | 1997-09-25 | 1999-04-13 | Sharp Corp | 画像処理装置 |
JPH11196319A (ja) * | 1997-12-26 | 1999-07-21 | Minolta Co Ltd | 撮像装置 |
JP2002042109A (ja) * | 2000-07-21 | 2002-02-08 | Topcon Corp | 医用画像合成処理装置及び方法並びに記憶媒体 |
JP2006119730A (ja) * | 2004-10-19 | 2006-05-11 | Seiko Epson Corp | 画像のつなぎ合わせ |
JP2006345400A (ja) * | 2005-06-10 | 2006-12-21 | Matsushita Electric Ind Co Ltd | ビデオカメラ装置 |
JP2009033224A (ja) * | 2007-07-24 | 2009-02-12 | Nippon Hoso Kyokai <Nhk> | 合成画像生成装置および合成画像生成プログラム |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140016866A1 (en) * | 2012-07-10 | 2014-01-16 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image |
US9224190B2 (en) * | 2012-07-10 | 2015-12-29 | Samsung Electronics Co., Ltd. | Method and apparatus for processing image taken under low illumination environment |
Also Published As
Publication number | Publication date |
---|---|
EP2613290B1 (en) | 2019-05-08 |
JPWO2012056518A1 (ja) | 2014-02-24 |
US8538192B2 (en) | 2013-09-17 |
CN102741878A (zh) | 2012-10-17 |
EP2613290A4 (en) | 2013-10-09 |
CN102741878B (zh) | 2015-08-19 |
JP4941950B1 (ja) | 2012-05-30 |
EP2613290A1 (en) | 2013-07-10 |
US20120177302A1 (en) | 2012-07-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP4941950B1 (ja) | 画像処理装置、画像処理方法及び画像処理プログラム | |
US8233055B2 (en) | Image processing device and program | |
CN107256532B (zh) | 图像处理装置、图像处理方法以及记录介质 | |
JP6100089B2 (ja) | 画像処理装置、画像処理方法およびプログラム | |
JP5687553B2 (ja) | 画像合成装置、画像合成方法及び画像合成プログラム | |
JP6682559B2 (ja) | 画像処理装置、画像処理方法、画像処理プログラム及び記憶媒体 | |
JP2007226643A (ja) | 画像処理装置 | |
US8861846B2 (en) | Image processing apparatus, image processing method, and program for performing superimposition on raw image or full color image | |
JP5493112B2 (ja) | 画像処理装置、画像処理方法及び画像処理プログラム | |
JP2010224607A (ja) | 画像合成装置及びプログラム | |
JP2010034964A (ja) | 画像合成装置、画像合成方法及び画像合成プログラム | |
US10095919B2 (en) | Image processing apparatus, image processing method and storage medium to suitably clip a subject region from a moving image | |
JP5158262B2 (ja) | 画像処理方法および画像処理装置 | |
JP5022498B2 (ja) | 画像処理装置、画像処理方法及び画像処理プログラム | |
JP4714038B2 (ja) | 画像の高解像度化方法及び装置 | |
US11195247B1 (en) | Camera motion aware local tone mapping | |
JP2005071223A (ja) | 多重画像作成装置、多重画像作成方法、多重画像作成プログラム及びプログラム記録媒体 | |
JP2016208083A (ja) | 画像処理装置、画像処理方法、及びプログラム | |
JP2014086097A (ja) | 画像処理装置、画像処理方法、画像処理プログラム及び記録媒体 | |
JP2007020087A (ja) | 撮像装置及び撮像方法 | |
JP2006119728A (ja) | 合成画像の表示 | |
JP5381207B2 (ja) | 画像合成装置及びプログラム | |
JP2011248475A (ja) | 画像処理装置、画像処理方法 | |
JP2009278655A (ja) | 画像処理装置、画像処理方法、及びデジタルカメラ | |
JP2015177529A (ja) | 画像処理装置、画像処理方法及びプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201080014815.5 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010542863 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13260278 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2010848135 Country of ref document: EP |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 10848135 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |