JP2010056771A - Device and method for reading image, program, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for reading an image having a simple device structure, rapidly reading an image, and acquiring a high-resolution image, and to provide a program, and a storage medium. <P>SOLUTION: This device for reading an image includes: an imaging means 11 to image a document at every predetermined cycle time; an evaluation means 41 to evaluate quality of image data picked up at every predetermined cycle time by the imaging means; an image data storage means 28 to store respective image data in association with an evaluation value obtained by the evaluation of the evaluation means 41; an instruction input means 16 to input an image acquisition instruction from a user; a pixel correlation means 45 to read image data having an evaluation value not smaller than a predetermined value, and detect a pixel of second image data obtained by imaging the same document corresponding to a pixel of first image data; and a pixel value storage means 46 to respectively store pixel values of a plurality of pixels associated with one another by the pixel correlation means in a pixel range of composite image data for which the number of pixels larger than that of the image data is secured when the image acquisition instruction is input. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image reading apparatus that acquires image data of an original, and more particularly to an image reading apparatus, an image reading method, a program, and a storage medium that place an original reading surface upward.

In a general scanner, an original is placed with the original surface facing down, and the original surface is in contact with the contact glass, and the original is optically scanned through the contact glass. For this reason, when scanning the document surface of the book, it is necessary to lift the book and turn the page each time a spread page is taken in, and place the book on the contact glass again, which is extremely troublesome for the user. Also, since the user cannot see the document surface while scanning the document, problems such as scanning an erroneous page or capturing an image when the document is not fully opened are likely to occur. These problems also occur when scanning a multi-page document with a scanner not equipped with an automatic document feeder (ADF).
On the other hand, various methods have been conceived for reading an original using an area sensor such as a digital camera instead of a scanner in a state where the reading surface faces upward as in a document camera, for example. For example, a reading method is conceivable in which a plurality of images taken by shifting a camera or subject by a minute distance are combined (for example, see Patent Document 1).

  Japanese Patent Laid-Open No. 2004-228688 includes an imaging camera and a movable document table for the purpose of making it possible to capture an image that exceeds the imaging area of the camera and converting a non-planar document into a plane, while moving the document table. An image reading system for photographing a document and synthesizing the obtained images is disclosed. In general, such a reading method is used for scanning a document or the like in which the positional relationship between the subject and the camera is fixed with high resolution.

  In addition, a technique is known in which high resolution is achieved by interpolating between pixels by synthesizing images captured continuously in a short time (see, for example, Patent Document 2). Japanese Patent Application Laid-Open No. 2004-228561 mainly shoots a document from a plurality of angles using a camera attached to a swingable support column for the purpose of preventing the illumination device from being reflected, and synthesizes the obtained images. And a method are disclosed.

  Also, by moving the platform on which the subject is placed and the camera to shoot the partial area of the subject, and joining these partial images, a method to ensure a constant resolution even when the subject is larger than the resolution of the camera Is already known (see, for example, Patent Document 3).

Patent Document 3 discloses a line sensor in which an image reflected by a mirror disposed obliquely above a subject is installed in an optical system having a variable optical path for the purpose of reading the subject without performing image pasting. An apparatus and method for reading in US Pat.
Japanese Patent No. 3931107 JP 2006-279828 A Japanese Patent Laid-Open No. 2007-82005

  However, the method of moving the camera or subject by a minute distance as in the method described in Patent Document 1 requires extremely accurate alignment, which leads to complicated, large, and expensive equipment configuration. There was a problem. In addition, as in the method described in Patent Document 2, a method using a continuous image requires an appropriate amount of movement of the subject in the continuous image. If the subject is a paper document, the camera Since the positional relationship with the subject does not change, the effect of this method cannot be expected.

Further, in the method described in Patent Document 1 or 2, since the partial areas are photographed and combined while moving the document table, a plurality of photographings are required to obtain one image, and the physical operation of the imaging system is performed. There is a problem that it takes time because it involves.
On the other hand, like the method described in Patent Document 3, the method of acquiring the image by changing the positional relationship between the camera and the subject and joining the partial images, or the method of shooting by changing the position and angle are described above. Although no problem occurs, it is necessary to move the imaging system to shoot, so that there is a problem that the apparatus configuration is complicated and it takes time to fetch a document once.
In view of the above problems, the present invention provides an image reading apparatus, an image reading method, a program, and a storage medium that read a document placed upward and acquire a high resolution image with a simple apparatus configuration. With the goal.

  In view of the above problems, the present invention provides an image reading apparatus having an original table on which an original is placed and an imaging unit that takes an image of an original placed on the original table with a reading surface substantially upward. An evaluation means for evaluating the quality of image data taken every cycle time, an image data storage means for storing each image data in association with the evaluation value evaluated by the evaluation means, and an image acquisition instruction by the user An instruction input unit, a pixel association unit that reads out image data having an evaluation value equal to or greater than a predetermined value, and that corresponds to a pixel of the first image data and detects from a pixel of the second image data obtained by photographing the same document; and an image Triggered by the input of the acquisition instruction, the pixel association unit associates each image data corresponding to the pixel range of the composite image data having a number of pixels larger than the number of pixels of the image data. And having a pixel value storage means for storing the pixel value of the number of pixels.

  It is possible to provide an image reading apparatus, a program, and a storage medium that have a simple apparatus configuration, read an image at high speed, and acquire a high-resolution image.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings.

  FIG. 1 schematically illustrates an example of image reading by the image reading apparatus 100. First, the camera 11 captures the document 12 every cycle time. The relative position of the document 12 with respect to the camera 11 changes little by little due to the placement, page turning by the user, or the self-healing force after the page turning. Since the camera 11 shoots the document 12 at every cycle time, the document 12 with the relative position slightly changed can be photographed. Therefore, even if the same document 12 is photographed, each element of the camera 11 is captured. Different minute portions (document pixels described later captured by one pixel of the camera) in the document 12 are incident and photographed.

  Each of the image data is taken with the same number of pixels G1, and the letter “A” is taken in the image data. Even if the same character “A” is photographed in the image data, the photographing range of each of the image data A to C is different because the document 12 is photographed while moving little by little. For this reason, the image reading apparatus 100 aligns the pixels of the plurality of image data with respect to the reference image data. That is, for each pixel of each image data, a pixel in which the same minute portion (original pixel) of the original is photographed is associated.

  In addition, the image reading apparatus 100 prepares an image composition field (memory) in which image data is vertically and horizontally several times (three times in the figure). Immediately after the preparation, nothing is stored in each pixel of the image composition field. For example, when the image data is tripled vertically and horizontally, the number of pixels is 9 × G1 pixels. The image reading apparatus 100 stores the pixels of each image data in the image composition field. In this case, since the image composition field is nine times as wide as the image data, in principle, all image data can be stored up to nine image data.

  For example, the pixel value of a pixel of any image data A is stored in any pixel in the composition area of the image composition field determined by the ratio of the number of pixels of the image data A to C and the number of pixels in the image composition field. In the same procedure, the pixel values of the other pixels of the image data A are stored every two vertical and horizontal pixels in the image composition field, and a composition area having the same width is set for each pixel value. As a result, the pixel value of the image data A is stored at a rate of one for every nine pixels in the image composition field.

Then, pixel values from other image data B, C, etc. are stored in the composition area of the image composition field in which no pixel value is stored. Since the correspondence between the pixels of each image data is clear, the pixel value of the corresponding pixel is stored in an empty pixel in the “3 × 3” synthesis area. The storage position may be common to each synthesis area.
For example, if the image data A is stored in (1, 1), the pixel value of the corresponding pixel in the image data B is stored in (1, 2), and the pixel value of the corresponding pixel in the image data C is (1, 1). 3) and so on. In principle, if there are nine image data, all the pixels in the image composition field can be filled.

  When the image data is not photographed in a number that satisfies all the pixels of the image composition field, the pixel value is complemented and stored in the pixels of the image composition field in which no pixel value is stored. Various methods are known for interpolation, and any method may be used.

  Through the above processing, composite image data having a resolution of a predetermined multiple (here, 9 times) the number of pixels G1 of the camera 11 is obtained. The document 12 is photographed with the document surface facing upward, and the composite image data of each page is acquired one after another simply by the user turning the pages of the document 12. Therefore, it is necessary to lift the document 12 and turn it over. Absent. In addition, since it is not necessary to move the camera 11 or the base of the document 12, an increase in cost can be suppressed.

  FIG. 2 shows an example of a configuration diagram of the image reading system 200. The image reading system 200 includes an image reading apparatus 100, a network 18, and a receiving computer 19 that is a distribution destination of the composite image data.

  The image reading apparatus 100 further includes a document camera device 50 and a control unit 60 connected to the document camera device 50. The document camera apparatus 50 includes a camera 11, a light source 13, a column portion 14 provided with the camera 11 and the light source 13, a document table 15, and a reading switch 16. The camera 11 starts shooting by changing the image data and then continuously shoots every cycle time, and synthesizes one page of composite image data every time the reading switch 16 is pressed by the user.

  The camera 11 converts, for example, light incident on each pixel into electricity using a photoelectric conversion element such as CMOS (Complementary Metal Oxide Semiconductor) or CCD (Charge Coupled Device), and A / D converts it to a predetermined luminance gradation (for example, (256 gradations) image data is output. The camera 11 captures the document 12 every cycle time. The cycle time is, for example, several frames / second to 30 frames / second, but may be constant or set by the user.

  The light source 13 illuminates the document 12 with light having a wavelength of visible light using, for example, a fluorescent lamp or LED. You may irradiate the light of the wavelength which can be image | photographed also at night like near infrared light other than visible light.

  The network 18 is, for example, a LAN in the office, one department, one floor, etc., a WAN that connects geographically separated LANs such as between branches, or the Internet. The network 18 may be either wired or wireless.

  Since the control part 60 should just be a computer and can implement | achieve each means mentioned later, the name is PC (Personal Computer), PDA (Personal Data Assistance), a mobile telephone, PHS (Personal Handyphone System), a smart phone, etc. What is necessary is just a computer provided with CPU which performs a program.

  FIG. 3 shows an example of a hardware configuration diagram of the control unit 60. The CPU 21, the RAM 22, the ROM 23, the storage medium mounting unit 24, the NIC 25, the input device 26, the display control unit 27, and the HDD (Hard Disk Drive) 28 are connected to each other via a bus. Since the hardware configuration diagram of the receiving computer 19 is the same, the description thereof is omitted.

  The control unit 60 is connected to the document camera device 50 by, for example, USB (Universal Serial Bus), IEEE 1394, wireless USB, Bluetooth, or the like. When the camera 11 starts photographing, it receives image data photographed by the camera 11 one after another. To do. The image data is subjected to processing described later to generate composite image data.

  The CPU 21 provides various functions by reading an OS (Operating System) and application software from the HDD 28 and executing them, and comprehensively controls processing performed by the control unit 60. In addition, the image processing program 30 is executed to realize each function described later.

  The RAM 22 is a working memory (main storage memory) that temporarily stores the image processing program 30 and data necessary for processing when the CPU 21 executes the image processing program 30, and the ROM 23 starts a basic input output system (BIOS) and an OS. Programs and setting files for storage.

  The storage medium mounting unit 24 is configured so that the storage medium 29 can be attached and detached, and is used when the image processing program 30 and data recorded in the storage medium 29 are read and installed in the HDD 28 and when data is written. The storage medium 29 may be a non-volatile memory such as a CD, DVD, Blu-ray disc, SD card, Memory Stick (registered trademark), multimedia card, xD card, or the like.

  The NIC 25 is an interface for connecting to the network 18, for example, an Ethernet (registered trademark) card, and executes processing according to a protocol defined in the physical layer and data link layer of the OSI basic reference model.

  The input device 26 is a device for inputting various operation instructions from the user, such as a keyboard and a mouse. The display control unit 27 draws on a display 31 such as a liquid crystal with a predetermined resolution, number of colors, and the like based on screen information instructed by the program. For example, a GUI (Graphical User Interface) screen is formed, and various windows and data necessary for operation are displayed on the display 31.

  The HDD 28 may be a nonvolatile memory such as an SSD (Flash Solid State Drive), and stores an image processing program 30 in this embodiment. The image processing program 30 is stored in the storage medium 29 and distributed, or downloaded from a predetermined server connected via the network 18 and installed in the HDD 28.

  FIG. 4 shows an example of a functional configuration diagram of the control unit 60. Each block in FIG. 4 is realized by the CPU 21 executing the image processing program 30 or hardware such as an ASIC (Application Specific Integrated Circuit).

<Evaluation unit 41>
The evaluation unit 41 will be described. The evaluation unit 41 calculates an evaluation value for determining whether or not composite image data may be generated from the image data. If the shooting range of the document 12 with respect to the number of pixels is too small or the shooting range is shifted, it should not be used for the composite image data. Similarly, if the image data is out of focus, blurred, low in contrast, or dark, the image data is not captured under appropriate conditions and should not be used for composite image data. For this reason, the evaluation part 41 evaluates each image data based on such an evaluation item, and determines whether it can be used for synthetic | combination image data. The evaluation unit 41 quantifies the following evaluation items for each image data, for example. The evaluation value is higher as the numerical value is higher.
“Number of pixels in which the document was shot / number of pixels, shooting position”
“Sharpness (number of edges), contrast, average luminance value”
FIG. 5A shows an example in which the document 12 is photographed small, and FIG. 5B shows an example in which the photographing position is shifted. Since the document table 15 has a substantially constant luminance, such as a solid color or a single color, an area having a constant luminance in the image data can be detected as a region where the document table 15 is photographed. Therefore, “the total number of pixels of the image data—the number of pixels with substantially constant luminance” becomes “the number of pixels in which the document is photographed”, and “the number of pixels in which the document is photographed / number of pixels” can be calculated.

  Further, when the photographing position is shifted as shown in FIG. 5B, a region with a substantially constant luminance appears at the left end, and a region with a substantially constant luminance does not appear at the right end. As described above, in the left and right or upper and lower pixel regions, when a region with a constant luminance is present at a predetermined value or more in one side and there is no region with a constant luminance on the other side, the shooting position may be shifted.

  The sharpness is evaluated from the number and size of edges by detecting horizontal and vertical edges in the image data. The contrast is evaluated from a known method such as a bias of the pixel value histogram, and the average luminance value is evaluated from the average of the pixel values of each pixel of the image data. Hereinafter, a value obtained by weighting and summing the numerical values of the respective evaluation items is referred to as an evaluation value.

  Although the document 12 is photographed while moving little by little, it can be determined from this evaluation value whether image data suitable for composition is photographed.

  The evaluation value is stored in the HDD 28 in association with the image data regardless of whether the evaluation value is equal to or higher than a certain value. For example, if even one of the numerical values of each evaluation item does not satisfy the threshold value, it may be discarded without being stored in the HDD 28 as inappropriate for use in the composite image data. If the evaluation value does not satisfy the threshold value, it may be discarded without being stored in the HDD 28.

<Pixel correspondence>
First, the image correction unit 44 performs various image processes such as trapezoidal correction and contrast correction on each image data, and corrects the image data to image data suitable for synthesis. In particular, when the document 12 is in the book format, the image data is distorted near the binding portion, so that the distortion is corrected so that, for example, the outer edge of the document 12 is rectangular. Further, since the brightness changes near the binding portion and the edge, the luminance unevenness is corrected so that the entire document 12 has a uniform brightness.

  The pixel association unit 45 associates the pixels of each image data with the pixels of the image composition field. FIG. 6A is an example of a diagram schematically illustrating the correspondence between the image composition field and the pixels of the image data. For example, when there are three image data A to C, the same or close original pixel I (for example, the right end of the first image of the letter “A”) exists, but the original 12 is photographed while moving little by little. Therefore, the possibility that the document pixel I is shot at the same pixel position is low. For example, in the image data A, the original pixel I is photographed at the pixel (x1, y1), the original pixel I is photographed in the pixel (x2, y2) in the image data B, and (x3, y3) in the image data C. A document pixel I is photographed on the pixel.

  Since the same document pixel I is stored in the composition area of the same image composition field, the image data A can be composed. Therefore, the pixel association unit 45 converts the document pixels I of the image data A to C into the same composition area. Associate with.

  For this reason, first, the pixels of the document images A to C are associated with each other. The pixel values of the image data on which the same document 12 is photographed should match by shifting up, down, left and right. Therefore, for example, the degree of matching of the pixel values is determined while shifting the pixels of the image data B and C with respect to the image data having the best evaluation value (here, the image data A). The pixel association unit 45 calculates the square of the difference between the pixel values and calculates the sum (total). This is repeated while shifting the pixels one by one. When the total is the smallest, it is determined that each pixel of the image data B and C is associated with each pixel of the image data A.

  Alternatively, the matching position may be detected by comparing only the edge portions. The pixel association unit 45 can reduce the processing load by detecting edges in the vertical direction and the horizontal direction, and executing the same calculation only for edges that are greater than or equal to a predetermined value.

Further, when a high resolution image is created by combining a plurality of images, it is generally preferable to perform alignment at a level (subpixel level) smaller than one pixel. As a method for aligning in units of subpixels, for example, the following method is available.
1) Approximate the edges and contours of figures (characters, etc.) included in a certain area with a function, and find the position where the error between the approximate functions in the comparison target area is minimized between images 2) Included in a certain area The feature point (line end point, intersection point, etc.) of the figure (character etc.) to be obtained is obtained, and the position where the deviation of the feature point in the comparison target region is minimized between images is obtained. FIG. FIG. 7B is an example of a diagram schematically showing the method 2). When the position of the image B is aligned with the image A in FIG. 7A, the images A and B are approximated by functions, respectively, and the position where the shift between the functions is minimized is determined while shifting the position of the image B.

  In FIG. 7B, the feature points of the image A and the image B are extracted from the original image data, respectively, and the position where the shift between the feature points is minimized is determined.

  Using such a method, alignment between images is strictly performed, and a new pixel is arranged at a position on the pixel range matching the shift. For example, when the resolution is increased to three times the original image, a shift of 1/3 pixel at the original resolution corresponds to a shift of one pixel at the high resolution. Therefore, when it is determined that there is a shift of 1/3 pixel at the time of alignment, a pixel is arranged adjacent to one pixel in the pixel range of the high resolution image. In reality, it is extremely rare to find a pixel that fits exactly within a pixel range in a high-resolution image, so keep the information about the pixels that did not fit, Used when complementing pixel values.

  Next, the pixel storage unit 46 stores each pixel of the image data A in the composition area of the image composition field. For example, when the composite image data has a size that is three times the length and width of the image data, the coordinates of (x1, y1) where the document pixel I was photographed are tripled in the x direction and y direction, respectively (3 × x1, 3 In (x1), the pixel value (x1, y1) of the image data A is stored.

  A plurality of pixels around this (3 × x1, 3 × y1) pixel is a synthesis region. In the figure, (3 × x1, 3 × y1) is the upper left, the rectangular region of 2 pixels on the right, and 2 pixels on the lower side is the synthesis region. (3 × x1, 3 × y1) may be associated with the center pixel of the 9 × 9 composite region, or may be associated with the upper right, lower right, and lower left pixels.

  Then, the pixel storage unit 46 corresponds to the pixel (x1, y1) of the image data A, the pixel value of the pixel (x2, y2) of the image data B, the pixel value of the pixel (x3, y3) of the image data C, Is stored in the synthesis area.

There are several methods for determining where to store the pixels of each image data in the synthesis area.
-Based on the result of aligning the image data with the best evaluation value with other image data, fill the pixels where the pixel value of the composite area is not stored.-Store the image data with a close evaluation value nearby. Considering the moving direction of the document 12 Stored in a predetermined order or at random When using the result of aligning the image data A with the best evaluation value and the other image data B, for example, move one pixel to the right If it corresponds in the positional relationship, the pixel value of the image data B is stored on the right side of the pixel of the image data A in the synthesis area. If the desired pixel value is already stored, the pixel value is then preferably stored in the pixel based on the alignment result.

  In addition, “image data with similar evaluation values” may be similar in appearance to the image data, and it is considered that discomfort in the composite image data can be reduced by arranging the image data close to each other. Therefore, if the image data A has the highest evaluation value, the pixel storage unit 46 refers to the evaluation value and determines the pixel value of the image data (for example, image data B) having the next highest evaluation value as the image in the synthesis area. The pixel value of the image data (for example, image data C) having the next highest evaluation value is stored in the pixel below the image data A in the synthesis area.

  Further, the moving direction of the document 12 may be taken into consideration. Since the document 12 is photographed every cycle time while moving little by little, the relative position for storing each pixel value in the synthesis area is determined in the direction in which the image data B and C are moved with respect to the image data A. For example, when the image data B moves in the upper left direction with respect to the image data A, the pixel value of the pixel (x1, y1) of the image data A is arranged on the lower right, and the pixel (x2, y2) of the image data B is placed on the upper left. ) And the pixel value of the pixel (x3, y3) of the image data C is stored at the upper left. Therefore, in this case, the composite region is not a rectangle but a linear region that is continuous in the upper left direction. By combining in consideration of the moving direction, spatially close pixels of the document 12 can be continuously arranged. In this case, since the aspect ratio may be different between the image data and the composite image data, correction is appropriately made so that the aspect ratios coincide with each other, and pixels without pixel values are filled by interpolation of the pixels described below.

<Pixel interpolation>
Next, pixel interpolation will be described. FIGS. 8A to 8D are examples of diagrams schematically illustrating pixel interpolation. 8A to 8D, pixel values are indicated by numerical values in the left composite region, and pixel values are indicated by grayscale shading in the right region. FIG. 8A shows an empty composite area in which no pixel value is stored, FIG. 8B shows a composite area in which the pixel value of the reference image data A is stored, and FIG. Each of the combined areas in which pixel values of image data other than image data A such as image data B and C are stored is shown.

  As a result, pixel values are not stored in 4 pixels in FIG. Such a state occurs when a sufficient number of image data satisfying the evaluation value is not captured. When an empty pixel exists in the synthesis area, the pixel interpolation unit 47 interpolates the empty pixel using another pixel value stored in the synthesis area. For example, the interpolation of pixel values may be linear interpolation or may be complemented using a quadratic expression or a cubic expression.

  In FIG. 8C, since the pixel of (2, 1) is empty, the pixel value of (2, 1) is “96” when linear interpolation is performed using the pixel values of pixels (1, 1) and (3, 1). " Similarly, since the pixel of (2, 3) is empty, the pixel value of (2, 3) becomes “32” when linear interpolation is performed using the pixel values of pixels (1, 3) and (3, 3). . Similarly, since the pixel of (3, 2) is empty, the pixel value of (3, 2) becomes “32” when linear interpolation is performed using the pixel values of pixels (3, 1) and (3, 3). . Similarly, since the pixel of (2, 2) is empty, the pixel value of (2, 2) becomes “64” when linear interpolation is performed using the pixel values of pixels (1, 1) and (3, 3). .

  By interpolation, the pixel values of each pixel in the synthesis area can be complemented and the pixel values can be stored in all the pixels. Although an example of interpolation is shown in FIG. 8C, extrapolation may be performed, and instead of interpolation by linear (primary expression), quadratic expression, cubic expression, or the like, for example, image data A to C May be stored in the pixel value of the empty pixel, or the pixel value of each pixel may be weighted with the evaluation value of the image data A to C to obtain the pixel value of the empty pixel. It may be stored.

  Further, in FIG. 8C, although complemented in the synthesis area, the adjacent synthesis area (not shown) corresponds to an adjacent pixel in the document 12, so that the pixel values of the pixels in the adjacent synthesis area are close. A value may be stored. For this reason, the pixel interpolation unit 47 may interpolate the pixel value of the pixel of interest using a pixel with an empty pixel value as the pixel of interest and using the surrounding (adjacent 8 pixels) pixel values. In this case as well, an average value of pixels around the target pixel may be used, or a smoothing filter such as a Gaussian filter may be used.

<Difference detection>
The difference detection unit 42 detects a pixel value difference between the image data for each image data. That is, a difference (which may be an absolute value or a square of the difference) obtained by subtracting pixel values of different image data captured at the same pixel position is calculated for each pixel, and a total of all the pixels is calculated. For example, when no document 12 is placed on the document table 15, the document table 15 is continuously photographed for each image data, so the difference between the image data is extremely small. On the other hand, when the document 12 is placed on the document table 15 or the page is turned, the difference changes greatly. Therefore, by detecting the difference, it is possible to detect that the document 12 is placed or the page is turned.

  Then, the deletion unit 43 deletes image data whose image data difference is within a predetermined value and image data whose difference is extremely large from the HDD 28. Similar image data can be prevented from squeezing the HDD 28 by the former, and the change of the image data is large and difficult to use for composition as in the case where the user turns the next page by the latter (while turning). Image data can be eliminated.

[Operation Procedure of Image Reading Apparatus 100]
The operation procedure of the image reading apparatus 100 will be described with reference to the flowchart of FIG. First, the user places the document 12 as a subject at a predetermined position on the document table 15. The difference detection unit 42 determines whether or not the user is about to place the document 12 on the document table 15 based on the change in the image data reflected on the camera 11 (S301). If the user does not intend to place the document 12 on the document table 15 or has not placed it (No in S301), the determination in step S301 is repeated. In this case, the image data captured by the camera 11 is discarded without being stored in the HDD 28.

  Note that it is desirable that the shooting range of the camera 11 be sufficiently wider than the document 12. As a result, even if the document 12 moves little by little, the entire surface of the document 12 can be captured with a larger amount of image data, which is advantageous when the synthesized image data is synthesized.

  When the user is about to place the document 12 on the document table 15 or places the document 12 (Yes in S301), the camera 11 starts to acquire image data necessary for generating the composite image data that is the final product. . As described above, the camera 11 acquires image data every predetermined cycle time (S302).

  Image data captured by the camera 11 at a predetermined time interval is stored in the HDD 28 of the control unit 60 every time it is captured (S303).

Next, the evaluation unit 41 evaluates the captured image data (S304). As described above, the evaluation is performed for each evaluation item such as whether the area of the document 12 in the image data is sufficient or whether the image data is sufficiently clear. The evaluation unit 41 stores the evaluation value in the HDD 28 in association with the image data (S305).
Next, the evaluation unit 41 determines whether or not the evaluation value is greater than or equal to a certain value (S306). If the evaluation value is not a certain value (No in S306), the evaluation unit 41 deletes the image data (S307). Thereby, it is possible to prevent generation of composite image data using image data having a low evaluation value.

  When the number of image data stored in the HDD 28 exceeds a certain number, it is desirable to delete the image data in order from the lowest evaluation value. As a result, the capacity of the HDD 28 can be saved.

  As described above, the image data is continuously acquired together with the evaluation value while the user moves the document 12 on the document table 15 or turns the page. That is, since the evaluation value is stored together with the image data, image data suitable for composition can be used. It should be noted that when the page 12 is turned, if the document 12 undergoes a large change, the difference detection unit 42 detects it, so that it can be detected that the image data has been switched.

  Since the user depresses the reading switch 16 after opening a page desired to read the document 12, the image reading apparatus 100 detects an image reading instruction by depressing the reading switch 16 (S308). The page turning may be detected and an image reading instruction may be used.

  The image reading apparatus 100 that has detected the image reading instruction (Yes in S308) reads out a predetermined number of image data from the image data stored in the HDD 28, and sends it to the image correction unit 44. The image correction unit 44 executes various image processing including trapezoidal correction and contrast correction on each image data, and converts the image data into an image suitable for synthesis (S309).

  Next, the pixel associating unit 45 determines, for example, the reference image data having the highest evaluation value, and performs alignment by analyzing the relative positional relationship between the other image data and the reference image data. Further, the pixel storage unit 46 prepares an empty image composition field (memory), and stores the pixel value of each image data in each pixel of the image composition field (S310). When synthesizing a predetermined number of image data, the image reading apparatus 100 transmits the synthesized image data to the receiving computer 19 via the network 18 (S311).

  In this embodiment, the composite image data is transmitted to the receiving computer 19 via the network 18, but the composite image data is sent to an image forming apparatus such as a printer, copying machine, MFP (Multifunction Peripheral), etc. In this case, it can be used as a scanner of a copier.

  As described above, the image reading apparatus 100 according to the present embodiment performs shooting with the document surface facing upward, and therefore it is not necessary to lift the document 12 every time the user turns the page of the document 12. In addition, since the user 12 is photographing the document 12 before the user presses the reading switch 16, it is easy to photograph image data whose position changes little by little. Further, the user can obtain a high-resolution image without waiting for reading of the image data. The image reading apparatus 100 includes the camera 11, the light source 13, and the control unit 60, and it is not necessary to provide the document table 15 or the operation unit of the camera 11, so that an increase in cost can be suppressed.

  In the first embodiment, the image data is evaluated based only on the evaluation value. However, the evaluation value is difficult to stabilize, such as glossy paper, but the image data may be captured to the extent that it can be combined. Therefore, in the present embodiment, in addition to the evaluation value based on the image data, a new evaluation pattern 40 is added to the shooting range, and whether or not the image data is shot with satisfactory quality based on the shooting quality of the evaluation pattern 40. The image reading apparatus 100 that determines whether or not will be described. The functional configuration diagram in the present embodiment is the same as the functional configuration diagram of FIG. 4, and the hardware configuration diagram of the control unit 60 is the same as that of FIG.

  FIG. 10 is an example of a diagram that schematically illustrates an example of image data captured together with the evaluation pattern 40. The evaluation pattern 40 is preferably fixed at a predetermined position on the document table 15. Thereby, the detection of the evaluation pattern 40 becomes easy. The evaluation pattern 40 is divided into squares, and predetermined shades or colors are arranged in the squares. In order to facilitate the edge detection, the outline that divides the square is black with a pixel value of zero or white with a pixel value of 255. Therefore, for example, when the evaluation pattern 40 is detected based on the outline of the square and the evaluation pattern 40 is detected, whether the pixel value of each square matches the pixel value of the master image registered in advance (whether the error is within the range). Or not). If they match, it can be determined that the image data has been properly captured.

  Therefore, in this embodiment, it is possible to detect whether or not image data that can be combined has been acquired from both the evaluation value detected from the image data and the photographing quality of the evaluation pattern 40. That is, in the image reading apparatus 100 that shoots the document 12 while turning pages, it is determined by using the static evaluation pattern 40 whether or not the image data of the dynamically changing document 12 has been acquired to the extent that it can be synthesized. It becomes possible.

FIG. 11 shows an example of a flowchart showing the operation procedure of the image reading apparatus 100 of the present embodiment. Since the overall procedure is the same as that of the first embodiment, only different portions are shown in FIG. As shown in FIG. 11, only steps S304 and S306 among steps S304 to S307 of FIG. 9 will be described.
・ S304
The evaluation unit 41 evaluates the image data stored in the HDD 28 and calculates an evaluation value. The calculation method has been described in the first embodiment. The evaluation unit 41 cuts out the evaluation pattern 40 from the image data. Since the photographing position of the evaluation pattern 40 may be substantially the same between the image data, the evaluation pattern 40 is detected from a preset range (for example, a predetermined range at the upper right of the image data). The evaluation unit 41 reads the standard template of the evaluation pattern 40, evaluates the difference between the pixel values while shifting the pixels within the set range, and detects that the evaluation pattern 40 is photographed at the position where the pixel values are the same. . Assume that the evaluation pattern 40 is captured in the pixel area at that time, and the evaluation pattern 40 is cut out.

  Next, the evaluation unit 41 detects vertical and horizontal edges from the photographed evaluation pattern 40 and determines whether or not a predetermined number of edges or more are detected. If a predetermined number or more of edges are detected, the image data may be captured with an appropriate focus.

  Further, the evaluation unit 41 creates a histogram for every 10 pixel values from the evaluation pattern 40, for example, and determines whether or not the pixel values are distributed in a predetermined standard distribution. For example, if each histogram is within a plus or minus threshold with respect to the standard distribution, it can be determined that the image is shot with an appropriate contrast. Further, when each histogram is moved to high pixels or low pixels as a whole with respect to the standard distribution, the standard distribution can be obtained by adjusting the brightness of the entire image data (for example, the influence of the shutter speed and the light source 13). Therefore, it may be determined that the image is captured with an appropriate contrast.

  Note that it is preferable to digitize the photographing quality of the evaluation pattern 40 in accordance with the number of edges and the degree of coincidence with the standard distribution of the histogram. By digitizing, the evaluation value and the shooting quality of the standard pattern 40 can be handled equally.

・ S306
Next, the evaluation unit 41 determines whether or not at least one of “Is the evaluation value equal to or greater than a certain value” or “Is the shooting quality of the evaluation pattern satisfy the criterion” is satisfied (S306)? The procedure when either one is satisfied (Yes in S306) is the same as that in the first embodiment.

  According to the image reading apparatus 100 of the present embodiment, by using the common evaluation pattern 40 for the image data, it is possible to reduce the extent to which the document type affects the determination of whether or not the image data used for the synthesis has been acquired. Can do. As a result, more accurate image quality evaluation can be performed, and high-quality composite image data can be generated even if the evaluation value varies slightly.

  In the first or second embodiment, it has been described that when the number of image data stored in the HDD 28 exceeds a certain number, the image data with the lowest evaluation value is deleted in order, but this embodiment specifically realizes this. The image reading apparatus 100 will be described.

  The image reading apparatus 100 continuously shoots with the document 12 moving little by little, but in many cases, the image reading apparatus 100 eventually stops. In a stationary state, the difference between the image data is almost constant. Therefore, in this embodiment, when the difference between the image data falls within a certain range, the image data with a low evaluation value is deleted.

  The functional configuration diagram in the present embodiment is the same as the functional configuration diagram of FIG. 4, and the hardware configuration diagram of the control unit 60 is the same as that of FIG.

  FIG. 12 shows an example of a flowchart showing a procedure for reading an image by the image reading apparatus 100 of the present embodiment. In FIG. 12, the same steps as those in FIG. 9 are denoted by the same reference numerals, and the description thereof is omitted.

  As described above, the image data captured by the camera 11 that satisfies the evaluation value or the imaging quality is stored in the HDD 28 in association with the evaluation value (S301 to S307).

  The evaluation unit 41 reads, from the HDD 28, image data having a high evaluation value among image data captured before the image data with respect to image data captured repeatedly (S401). Then, the difference detection unit 42 calculates the difference between the image data captured at that time and the read image data (S402).

  The deletion unit 43 determines whether or not the difference is within a certain range (S403). If the difference is within the certain range (including zero and extremely small), the document 12 is photographed and synthesized in a substantially completely stationary state. Even so, the improvement in resolution is not expected. Accordingly, the deletion unit 43 deletes the lower evaluation value from the HDD 28 while leaving the higher evaluation value among the two image data to be compared (S404).

  The evaluation unit 41 and the deletion unit 43 repeat steps S402 to S403 for the stored image data (S405). Thereby, only the image data with a small difference can be left with a high evaluation value.

  If the difference is extremely large, it can be determined that the document 12 has changed significantly, and it is determined that the captured image data cannot be combined and is deleted from the HDD 28. As a result, even if the user moves the document 12 or turns the next page by mistake, the quality of the composite image data can be prevented from deteriorating.

  The subsequent steps are the same as those in the first or second embodiment, and when the image reading instruction is detected, the combined image data is combined.

  According to the image reading apparatus 100 of the present embodiment, since the image data having a certain difference is deleted, it is possible to prevent only the same image data from being stored in the storage unit. Further, by appropriately deleting image data that is not suitable for composition, only image data suitable for image composition can be used for composition.

It is an example of the figure which illustrates typically the reading of the image by an image reader. It is an example of a block diagram of an image reading system. It is an example of the hardware block diagram of a control part. It is an example of the function block diagram of a control part. FIG. 3 is an example of a diagram schematically illustrating a relationship between a document and a shooting range. It is an example of the figure which illustrates typically matching with the pixel of an image composition field and image data. It is an example of the figure which illustrates the method of aligning in a subpixel unit typically. It is an example of the figure which illustrates pixel interpolation typically. It is an example of the flowchart figure which shows the operation | movement procedure of an image reading apparatus. It is an example of the figure which illustrates typically an example of the image data image | photographed with the evaluation pattern. FIG. 10 is an example of a flowchart illustrating an operation procedure of the image reading apparatus (second embodiment). FIG. 10 is an example of a flowchart illustrating an operation procedure of the image reading apparatus (third embodiment).

Explanation of symbols

11 Camera 12 Document 13 Light source 14 Column
15 Document Plate 16 Reading Switch 18 Network 19 Receiving Computer 21 CPU
28 HDD
DESCRIPTION OF SYMBOLS 30 Image processing program 40 Evaluation pattern 41 Evaluation part 42 Difference detection part 43 Deletion part 44 Image correction part 45 Pixel matching part 46 Pixel storage part 47 Pixel interpolation part 50 Document camera apparatus 60 Control part 100 Image reading apparatus 200 Image reading system

Claims (12)

An image reading apparatus for reading a document placed on a platen with a reading surface substantially upward,
Photographing means for photographing a document every predetermined cycle time;
Evaluation means for evaluating the quality of image data taken by the photographing means for each cycle time;
Image data storage means for storing each image data in association with the evaluation value evaluated by the evaluation means;
An instruction input means for inputting an image acquisition instruction by the user;
Pixel association means for reading out image data having an evaluation value equal to or greater than a predetermined value, and detecting pixels of second image data obtained by photographing the same document corresponding to pixels of the first image data;
Triggered by the input of the image acquisition instruction,
The composite image data in which the number of pixels larger than the number of pixels of the image data is secured, and the pixel correspondence unit associates the pixel range of the composite image data determined by the number of pixels of the image data and the composite image data with each other. Pixel value storage means for storing pixel values of a plurality of pixels of each image data,
An image reading apparatus comprising: When pixel values of image data are not stored in all of the pixel range, an interpolation unit that interpolates pixel values of pixels in which the pixel value is not stored among pixels of the composite image data with pixel values of surrounding pixels,
The image reading apparatus according to claim 1, further comprising: The evaluation means evaluates the quality of the image data based on at least one of a ratio of the number of pixels in which the document is photographed to the number of pixels in the photographing means or a position in which the document is photographed in the photographing range of the photographing means. To
The image reading apparatus according to claim 1, wherein the image reading apparatus is an image reading apparatus. The evaluation means evaluates the quality of the image data based on at least one of the sharpness, contrast or average luminance value of the image data;
The image reading apparatus according to claim 1, wherein the image reading apparatus is an image reading apparatus. An evaluation pattern with shading is arranged within the photographing range of the photographing means,
The evaluation means evaluates the quality of image data based on whether or not a difference between a pixel value distribution of a photographed evaluation pattern and a standard pixel value distribution of the evaluation pattern stored in advance is within a predetermined range. To
The image reading apparatus according to claim 1, wherein the image reading apparatus is an image reading apparatus. Deleting means for deleting image data having an evaluation value equal to or less than a predetermined value from the image data storage means;
The image reading apparatus according to claim 1, comprising: A difference detection unit that detects a difference between pixel values of image data captured by the imaging unit and image data stored in the image data storage unit;
The deletion means deletes image data with a lower evaluation value when the difference is within a predetermined value.
The image reading apparatus according to claim 6. The pixel value storage means includes
For the pixel storing the pixel value of the image data having the best evaluation value in the pixel range,
The pixel storing means determines the pixel for storing the pixel value of the other image data so that the correspondence relationship between the image data having the best evaluation value and the pixel of the other image data can be obtained. To
The image reading apparatus according to claim 1. When the horizontal length or vertical length of the composite image data is n times the image data,
The pixel value storage means sets the pixel range to a pixel position in the composite image data obtained by multiplying the pixel positions of any of a plurality of pixels associated with each other by n times vertically and horizontally. Set,
The image reading apparatus according to claim 1. An image reading method for reading a document placed on a platen with a reading surface substantially upward,
A step of photographing means for photographing a document every predetermined cycle time;
An evaluation unit that evaluates the quality of image data captured by the imaging unit every predetermined cycle time;
Storing each image data in the image data storage means in association with the evaluation value evaluated by the evaluation means;
An instruction input means for inputting an image acquisition instruction by a user;
Reading out image data having an evaluation value equal to or greater than a predetermined value, and detecting a pixel in the second image data obtained by photographing the same document corresponding to the pixel in the first image data.
Pixel value storage means triggered by the input of the image acquisition instruction,
The composite image data in which the number of pixels larger than the number of pixels of the image data is secured, and the pixel correspondence unit associates the pixel range of the composite image data determined by the number of pixels of the image data and the composite image data with each other. Storing pixel values of a plurality of pixels of each image data,
An image reading method comprising: Connected to a document camera having an original table on which an original is placed, a photographing means for photographing an original placed on the original table with a reading surface substantially upward, and an instruction input means for inputting an image acquisition instruction by a user Computer
Evaluation means for evaluating the quality of image data taken by the photographing means every predetermined cycle time;
Image data storage means for storing each image data in association with the evaluation value evaluated by the evaluation means;
Pixel association means for reading out image data having an evaluation value equal to or greater than a predetermined value, and detecting pixels of second image data obtained by photographing the same document corresponding to pixels of the first image data;
Triggered by the input of the image acquisition instruction,
The composite image data in which the number of pixels larger than the number of pixels of the image data is secured, and the pixel correspondence unit associates the pixel range of the composite image data determined by the number of pixels of the image data and the composite image data with each other. Pixel value storage means for storing pixel values of a plurality of pixels of each image data,
A program characterized by functioning as   A computer-readable storage medium storing the program according to claim 11.

Images