JP2013123119A - Image processing apparatus, image forming apparatus, image reading apparatus, image processing method, computer program, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing apparatus, an image forming apparatus, an image reading apparatus, an image processing method, a computer program, and a recording medium for extracting a proper area of a document image even if a document has not been read properly.SOLUTION: In a case where an image forming apparatus 100 is instructed to execute a document inclination and size correction function, a color image input apparatus 1 reads a document in a maximum readable size. On the basis of image data acquired by the color image input apparatus 1, the document inclination and size detection unit 24 of a color image processing apparatus 2 detects a cut off area which has to be extracted as a document area. A document inclination and size correction unit 26 performs inclination and size correction using a correction parameter based on the cut off area detected by the document inclination and size detection unit 24.

Description

  The present invention relates to an image processing apparatus, an image forming apparatus, an image reading apparatus, an image processing method, and a computer program capable of extracting an appropriate original image area from image data obtained by reading an area larger than the original. And a recording medium.

  There has been proposed an image reading apparatus that reads an original with a maximum readable size, sets a paper size area set by a user in a crop box, and extracts the area in the crop box as read image data. (For example, patent document 1). Specifically, the system disclosed in Patent Document 1 detects a portion of an image data read from a document whose image density is equal to or higher than a preset threshold as a region where an image exists. Then, for example, an area having a shape and size suitable for the paper size set by the user is set in the crop box so that the area where the image exists is in the center of the crop box.

JP 2007-174479 A

However, in the method disclosed in Patent Document 1, for example, when the document is a photograph with a light background, the edge portion of the document may not be detected correctly. If the edge part of the document is not detected correctly, the crop box is set at the specified position based on the image data read from the document, and the crop box is set in the state where the photo area to which the crop box is originally set is missing. There is a risk of being.
Further, in order to set the crop box at the correct position, when the user manually performs the setting or fine adjustment of the crop box, there is a problem that the user is forced to perform complicated work.

  The present invention has been made in view of such circumstances, and an object of the present invention is to extract an appropriate area of an original image even when the original is not read in an appropriate state. An image processing apparatus, an image forming apparatus, an image reading apparatus, an image processing method, a computer program, and a recording medium are provided.

  An image processing apparatus according to the present invention includes a rectangular document, and extracts image data read from the document from image data obtained by reading an area larger than the document. An edge detection unit for detecting an edge of an image based on the image data based on the data; an original edge extraction unit for extracting edges on the four sides of the original document based on the edge detected by the edge detection unit; A reliability calculation unit that calculates the reliability of each edge extracted by the document edge extraction unit, a reception unit that receives the paper size of the output image, the reliability calculated by the reliability calculation unit, and the reception unit And an area specifying unit for specifying an extraction area to be extracted as the document image data based on the accepted paper size.

  According to the present invention, the image processing apparatus detects the edge of the image based on the image data, and extracts the edges on the four sides of the document based on the detected edge. The image processing apparatus calculates the reliability of each edge, and specifies the document image area (extraction area) based on the calculated reliability and the paper size of the output image received from the outside. Since the extraction region (cut region) is specified in accordance with the reliability of the four side edges extracted as each side of the document image, even if the document is difficult to extract all four side edges correctly, the extraction region Detection accuracy can be improved. Therefore, even when the original is not properly read, such as when it is difficult to detect the edge portion of the original correctly, appropriate original image data can be extracted. The document image data is image data read from a document or image data read from an image formed on a document, and the document image area is an area defined by the paper size of the document, or a document This is an area of an image formed in the area.

  The image processing apparatus according to the present invention includes a determination unit that determines whether or not the reliability of each edge calculated by the reliability calculation unit is greater than or equal to a predetermined threshold, and the region specifying unit is When it is determined that at least one edge is equal to or greater than a predetermined threshold, the paper size extraction area received by the receiving unit is specified with reference to an edge having a reliability equal to or higher than the predetermined threshold. And

  According to the present invention, when at least one of the reliability of each edge on the four sides of the document is equal to or greater than a predetermined threshold, the paper size set from the outside with the reliability edge equal to or greater than the predetermined threshold as a reference Specify the extraction area. Since the extraction area is specified based on an edge whose reliability is equal to or higher than a predetermined value, an extraction area that is more approximate to the actual original image area can be detected even if it is difficult to correctly extract all four edges.

  The image processing apparatus according to the present invention includes a generation unit that generates display data for displaying an image obtained by superimposing information indicating the extraction region specified by the region specifying unit on an image based on the image data. It is characterized by.

  According to the present invention, for example, an image in which a frame indicating an extraction region to be extracted is superimposed on an image based on image data can be displayed. Therefore, it is possible to easily determine whether or not the extraction region is appropriate.

  An image processing apparatus according to the present invention includes: a display unit that displays an image based on display data generated by the generation unit; a change reception unit that receives a change in the extraction region based on the image displayed by the display unit; And an extraction region changing unit that changes the extraction region specified by the region specifying unit based on the change received by the change receiving unit.

  According to the present invention, for example, by displaying an image in which a frame indicating an extraction region to be extracted is superimposed on an image based on image data, the extraction region is notified to the user, and the extraction region is not appropriate. Accepts a change in the extraction area from the user and changes the extraction area. Therefore, when an appropriate extraction area is not specified (detected), the extraction area can be changed by the user, and an appropriate extraction area can be specified even for a document whose edges are difficult to extract correctly. .

  In the image processing apparatus according to the present invention, the edge detection unit is configured to detect edges on the four sides of the image, and for each pixel constituting the edge detected by the edge detection unit, each pixel of interest And a counting unit that counts the number of pixels in which the difference value of coordinate information between each pixel of interest and a pixel at a predetermined interval is less than a predetermined value as an effective count value for each pixel of interest, and for each edge, the counter unit A calculation unit that calculates an average value of the counted effective count values, and the document edge extraction unit calculates the average calculated by the calculation unit among the pixels constituting the edge on one side detected by the edge detection unit. A pixel having an effective count value greater than the value, the pixel at the most distant position from the other side facing the one side is extracted as an edge on one side of the document, and the other side detected by the edge detection unit Among the pixels constituting the edge of the document, a pixel having an effective count value larger than the average value calculated by the calculation unit, and a pixel at a position farthest from the one side is extracted as an edge on the other side of the document. Among the pixels constituting the edge on another side intersecting the one side detected by the edge detection unit, the pixel having an effective count value larger than the average value calculated by the calculation unit, and the other side The pixel at the position farthest from the other side opposite to the other side is extracted as the edge on the other side, and among the pixels constituting the edge on the other side detected by the edge detection unit, The pixel having an effective count value larger than the average value calculated by the calculation unit, the pixel at the position farthest from the other side is extracted as an edge on the other side, and the reliability calculation Department is front Wherein the document edge extraction unit are the valid count values of pixels extracted as the edge so that the reliability of each edge.

  According to the present invention, for each pixel detected as an edge on each side of the image, the number of pixels for which the difference value of the coordinate information between each pixel of interest and each pixel from the pixel of interest is less than a predetermined value is calculated. Then, counting is performed as an effective count value for each pixel of interest, and an average value of the effective count values is calculated for each edge. Then, for each edge, among the pixels extracted as each edge, a pixel having an effective count value larger than the average value, and a pixel at a position where the area surrounded by the four edges is further expanded is set to each side of the document. The side edge. Further, the effective count value of the pixel extracted as each edge is set as the reliability of each edge. Therefore, a wider area can be specified as the extraction area, and the image can be prevented from being lost when the extraction area is extracted.

  An image forming apparatus according to the present invention includes any one of the above-described image processing apparatuses and an image forming unit that forms an image based on image data processed by the image processing apparatus.

  According to the present invention, it is possible to prevent the image output based on the extracted data from being lost by extracting the data of the document image based on the extraction area specified with high accuracy. Therefore, an appropriate image can be output.

  An image reading apparatus according to the present invention includes an image reading unit that reads a document and acquires image data, and any one of the image processing devices described above, and processes an image read by the image reading unit by the image processing device. It is made to do so.

  According to the present invention, the data in the extraction area specified with high accuracy is extracted from the image data obtained by reading an area larger than the original, so that the original cannot be read in an appropriate state. However, appropriate document image data can be extracted.

  In the image processing method according to the present invention, the image processing apparatus includes a rectangular document and extracts image data read from the document from image data obtained by reading an area larger than the document. In the method, the image processing device detects an edge of an image based on the image data based on the image data; and the image processing device detects four edges of the document based on the detected edge. A step of extracting each edge, a step in which the image processing apparatus calculates a reliability of each extracted edge, a reliability calculated by the image processing apparatus, and a paper size of the output image received from the outside And a step of specifying an extraction area to be extracted as the document image data.

  According to the present invention, even when the original is not properly read, such as when the edge portion of the original is difficult to be detected correctly, the same appropriate as when the original is read in an appropriate state. Document image data can be obtained.

The computer program according to the present invention is a computer program for causing a computer to extract document image data read from the document from image data obtained by reading a region wider than the document including a rectangular document. And causing the computer to detect an edge of an image based on the image data based on the image data, and causing the computer to extract edges on the four sides of the document based on the detected edge. A step of causing the computer to calculate the reliability of each extracted edge; and as the original image data based on the calculated reliability and the paper size of the output image received from the outside. Identifying an extraction region to be extracted. To.
According to the present invention, the above-described image processing apparatus can be realized on a computer by causing the computer program to be read and executed.

A recording medium according to the present invention stores the above-described computer program.
According to the present invention, the above-described image processing device can be realized on a computer by causing the computer to read and execute the computer program recorded on the recording medium.

  According to the present invention, an appropriate document image area can be extracted even when the document is not properly read, such as when the edge portion of the document is difficult to detect correctly. That is, even if the original image is difficult to extract correctly, the detection accuracy of the extraction area (original image area) can be improved.

1 is a block diagram illustrating a configuration of an image forming apparatus according to a first embodiment. It is a schematic diagram which shows the example of a setting screen displayed on an operation panel. FIG. 4 is a block diagram illustrating a configuration of a document inclination / size detection unit. It is a schematic diagram which shows a coordinate information table. It is a schematic diagram which shows a tangent angle correspondence table. 6 is a flowchart illustrating a processing procedure performed by a document area detection unit. 6 is a flowchart illustrating a processing procedure performed by a document area detection unit. It is a schematic diagram for demonstrating the determination process whether each edge is effective. It is a schematic diagram which shows the candidate of a left part largest effective edge and a right part largest effective edge. It is a schematic diagram for demonstrating the determination process of whether it is isolated point noise. It is a flowchart which shows the procedure of the process by a cutting area determination part. It is a schematic diagram which shows a cutting area. It is a schematic diagram which shows a cutting area. It is a flowchart which shows the procedure of the 5th cutting area determination process. It is a flowchart which shows the procedure of the 5th cutting area determination process. It is a schematic diagram which shows a cutting area. It is a schematic diagram for demonstrating the process by the parameter display part for area | region display. It is a schematic diagram which shows a cut area confirmation screen. It is a flowchart which shows the procedure of the confirmation and correction | amendment process of a cutting area. It is a schematic diagram which shows a setting screen. It is a schematic diagram for demonstrating an interpolation calculation. FIG. 4 is a block diagram illustrating a configuration of an image forming apparatus according to a second embodiment. FIG. 10 is a block diagram illustrating a configuration of an image reading apparatus according to a third embodiment. It is a schematic diagram which shows the structural example at the time of applying the image processing apparatus which concerns on this invention to a computer system.

  Hereinafter, an image processing apparatus, an image forming apparatus, an image reading apparatus, an image processing method, a computer program, and a recording medium according to the present invention will be described in detail with reference to the drawings illustrating embodiments thereof.

(Embodiment 1)
FIG. 1 is a block diagram illustrating a configuration of an image forming apparatus according to the first embodiment. An image forming apparatus 100 (for example, a digital color copying machine) includes a color image input device 1, a color image processing device 2 (image processing device), a color image output device 3 (image forming unit), and an operation panel for performing various operations. 4. An image memory 5 is provided. Image data of RGB (R: red, G: green, B: blue) analog signals obtained by reading a document with the color image input device 1 is output to the color image processing device 2, and the color image processing device 2. Then, a predetermined process is performed and output to the color image output device 3 as a digital color signal of CMYK (C: cyan, M: magenta, Y: yellow, K: black).

The color image input apparatus 1 is a scanner including a device that converts optical information into an electrical signal, such as a CCD (Charged Coupled Device), for example, and converts a reflected light image (reflectance signal) from an original image to an RGB analog. Read as a signal. The color image input device 1 outputs the read RGB signals to the color image processing device 2. Further, in the color image input apparatus 1, for example, the document presser base is configured by a white plate.
The color image output device 3 is an image forming unit using an electrophotographic method, an ink jet method, or the like that outputs image data of a document image onto a recording sheet, for example.

  The operation panel 4 is, for example, a touch panel type liquid crystal display, and includes an image display device such as a liquid crystal display, a setting button, and the like. The operation panel 4 displays an operation menu for operating the image forming apparatus 100, a preview image before image output, and the like. The operation panel 4 accepts operations performed on the image forming apparatus 100 by the user via setting buttons such as a numeric keypad for inputting numerical values and a start key for instructing start of operation. Based on information input from the operation panel 4, operations of the color image input device 1, the color image processing device 2, and the color image output device 3 are controlled. The operation panel 4 may be provided integrally with the image forming apparatus 100 or may be detachable from the image forming apparatus 100.

  The color image processing apparatus 2 includes an A / D (analog / digital) conversion unit 21, a shading correction unit 22, an input tone correction unit 23, a document inclination / size detection unit 24, a display control unit 25, and a document inclination / size correction unit. 26, an area separation processing unit 27, a color correction unit 28, a black generation and under color removal unit 29, a spatial filter processing unit 30, an output gradation correction unit 31, a gradation reproduction processing unit 32, and the like, and includes an ASIC (Application Specific Integrated Circuit). ) Etc.

  The A / D conversion unit 21 converts the RGB signal input from the color image input apparatus 1 into, for example, a 10-bit digital signal, and outputs the converted RGB signal to the shading correction unit 22.

  The shading correction unit 22 performs a correction process for removing various distortions generated in the illumination system, the imaging system, the imaging system, and the like of the color image input apparatus 1 on the input RGB signal. In addition, the shading correction unit 22 adjusts the color balance, performs a process of converting the RGB signal (RGB reflectance signal) into a density (pixel value) signal, and converts the processed RGB signal into an input tone correction unit. To 23.

  The input tone correction unit 23 performs image quality adjustment processing such as background density removal and contrast adjustment on the RGB signal from which various distortions have been removed by the shading correction unit 22. The input tone correction unit 23 outputs the processed RGB signals to the document inclination / size detection unit 24 and the document inclination / size correction unit 26.

The document inclination / size detection unit 24 extracts an edge of the document from the RGB signal whose input gradation has been corrected, and determines a cutting area (cutting area candidate) based on the extracted edge. The document inclination / size detection unit 24 displays the determined cut area on the operation panel 4 (display unit) via the display control unit 25 to notify the user.
The user determines whether or not the cut area displayed on the operation panel 4 is appropriate. If the cut area is not appropriate, the user performs fine adjustment via the operation panel 4.

When the user determines that the cut area displayed on the operation panel 4 is appropriate, the display control unit 25 notifies the document inclination / size detection unit 24 to that effect. Accordingly, the document inclination / size detection unit 24 performs correction parameters for the inclination / size correction processing (inclination angle, corrected paper size, correction processing start position, etc.) based on the information indicating the cut-out area. And the calculated correction parameters are output to the document inclination / size correction unit 26. If it is determined that the user is not appropriate and the user makes fine adjustments to the cutout area, the display control unit 25 displays information indicating the cutout area finely adjusted by the user as a document inclination / size detection unit. 24. In this case, the document inclination / size detection unit 24 calculates correction parameters for the inclination / size correction processing based on the information indicating the cut-out area notified from the display control unit 25, and the document inclination / size correction unit. 26.
The document inclination / size detection unit 24 and the display control unit 25 appropriately store image data to be processed and information necessary for various processes in the image memory 5. Details of specific processing performed by the document inclination / size detection unit 24 and the display control unit 25 will be described later.

  The document inclination / size correction unit 26 performs an inclination / size correction process on the RGB signal processed by the input tone correction unit 23 based on the correction parameters acquired from the document inclination / size detection unit 24. The processed RGB signal is output to the region separation processing unit 27. The details of specific processing performed by the document inclination / size correction unit 26 will be described later. In the present embodiment, when the execution of the “original skew / size correction” function is selected via the setting screen as shown in FIG. 2 as a function provided in the image forming apparatus 100, the original skew / size detection unit The correction parameter calculation process 24 and the inclination / size correction process by the document inclination / size correction unit 26 are executed.

  The region separation processing unit 27 separates each pixel in the input image into, for example, a character region, a dot region, a photographic region (continuous tone region), or the like based on the input RGB signal. . Based on the separation result, the region separation processing unit 27 sends a region identification signal indicating which region each pixel belongs to to the black generation and under color removal unit 29, the spatial filter processing unit 30, and the gradation reproduction processing unit 32. Output. The region separation processing unit 27 outputs the RGB signal input from the document inclination / size correction unit 26 to the subsequent color correction unit 28 as it is. When the document inclination / size correction unit 26 stores the image data (RGB signal) after the inclination / size correction processing in the image memory 5, the region separation processing unit 27 performs the image after the inclination / size correction processing. Data may be read from the image memory 5. Further, when the document inclination / size correction unit 26 does not perform the inclination / size correction processing, the RGB signal input from the input gradation correction unit 23 is output to the region separation processing unit 27 as it is. The area separation processing unit 27 outputs the RGB signal input from the input gradation correction unit 23 via the document inclination / size correction unit 26 to the color correction unit 28 as it is.

  The color correction unit 28 converts the input RGB signal into a CMY color space, performs color correction in accordance with the characteristics of the color image output device 3, and outputs the corrected CMY signal to the black generation and under color removal unit 29. To do. Specifically, the color correction unit 28 performs a process of removing color turbidity from the CMY signal based on the spectral characteristics of the CMY color material including the unnecessary absorption component in order to achieve faithful color reproduction.

  The black generation and lower color removal unit 29 performs black generation processing for generating a black (K) signal from the CMY signal, and K obtained by black generation from the original CMY signal for the CMY signal input from the color correction unit 28. A process of generating a new CMY signal by subtracting the signal is performed. As a result, the CMY3 color digital signals are converted into CMYK 4 color digital signals (hereinafter referred to as CMYK signals). The black generation and under color removal unit 29 outputs the CMYK signal obtained by converting the CMY signal to the spatial filter processing unit 30. The black generation and under color removal unit 29 performs processing according to the region indicated by the region identification signal.

  As an example of the black generation process performed by the black generation and under color removal unit 29, a method of generating black by skeleton black is used. In this method, the input / output characteristic of the skeleton curve is y = f (x), the input data is C, M, Y, the output data is C ′, M ′, Y ′, K ′, UCR (Under Color) Assuming that the (removal) rate is α (0 <α <1), the data output by the black generation and under color removal processing is K ′ = f (min (C, M, Y)) and C ′ = C−αK, respectively. ', M' = M-αK ', Y' = Y-αK '.

  Here, the UCR rate α (0 <α <1) indicates how much CMY is reduced by replacing the portion where CMY overlaps with K. The above equation shows that the K signal is generated in accordance with the smallest signal strength among the CMY signal strengths.

  The spatial filter processing unit 30 performs spatial filter processing on the image data of the CMYK signal input from the black generation and under color removal unit 29 using a digital filter corresponding to the region indicated by the region identification signal. As a result, the spatial frequency characteristic of the image data is corrected, and blurring or graininess deterioration of the output image in the color image output device 3 is prevented. For example, for a region separated into character regions by the region separation processing unit 27, the spatial filter processing unit 30 uses a filter with a large amount of high-frequency component enhancement in order to improve the reproducibility of black characters or color characters. Use for sharpening enhancement. Further, the spatial filter processing unit 30 performs low-pass filter processing for removing the input halftone dot component on the region separated into halftone dot regions by the region separation processing unit 27. The spatial filter processing unit 30 outputs the processed CMYK signal to the output tone correction unit 31.

  The output tone correction unit 31 performs output tone correction processing (γ correction processing) based on the characteristic value of the color image output device 3 for the CMYK signal input from the spatial filter processing unit 30 to correct output tone. The processed CMYK signal is output to the gradation reproduction processing unit 32.

  The gradation reproduction processing unit 32 performs halftone processing corresponding to the region indicated by the region identification signal on the image data of the CMYK signal input from the output gradation correction unit 31. For example, for the region separated into character regions by the region separation processing unit 27, the tone reproduction processing unit 32 reproduces high-frequency components in the color image output device 3 in order to improve the reproducibility of black characters or color characters. A binarization process or a multi-level process is performed using a high-resolution screen suitable for the above.

  In addition, the gradation reproduction processing unit 32 finally separates the image into pixels for the regions separated into halftone dot regions by the region separation processing unit 27 so that each gradation can be reproduced. Tone reproduction processing (halftone generation) is performed. Furthermore, the gradation reproduction processing unit 32 performs binarization processing or multi-value processing on the screen with an emphasis on gradation reproducibility in the color image output device 3 for the region separated into the photographic regions by the region separation processing unit 27. The process is performed.

  The color image processing apparatus 2 temporarily stores the image data (CMYK signal) processed by the gradation reproduction processing unit 32 in the image memory 5 and reads out the image data stored in the image memory 5 at a predetermined timing for image formation. The read image data is output to the color image output device 3.

  The image forming apparatus 100 includes a CPU (Central Processing Unit) or a DSP (Digital Signal Processor) (not shown), and various processes performed by the image forming apparatus 100 are controlled by the CPU or DSP.

FIG. 2 is a schematic diagram illustrating an example of a setting screen displayed on the operation panel 4. In the image forming apparatus 100 according to the first exemplary embodiment, when the user instructs the execution of the document tilt / size correction function via the setting screen as illustrated in FIG. Processing by the size correction unit 26 is performed. The setting screen shown on the upper side of FIG. 2 is a screen for instructing the setting of various information and the execution of each function regarding each function provided in the image forming apparatus 100. The screen shown at the bottom of FIG. 2 is a screen for setting information related to the document tilt / size correction function. When the “original skew / size correction” button in the screen shown in the upper part of FIG. 2 is operated, the screen shown in the lower part of FIG. 2 is displayed on the operation panel 4. Sets the paper size of the area cut out in the tilt / size correction process. In the example of the screen shown at the bottom of FIG. 2, L size size, 2L size size, KG size, quarter size, six size, and eight size are displayed as settable photo paper sizes. Although not shown, A5 size, A4 size, A3 size, B6 size, B5 size, B4 size, etc. are displayed as general paper sizes on the screen shown at the bottom of FIG.
The document inclination / size detection unit 24 determines a cut-out area candidate to be cut out from image data read from the document based on the paper size set (selected) in this way.

  As described above, in this embodiment, when the execution of the document inclination / size correction function is instructed via the setting screen shown in FIG. 2, the document inclination / size detection unit 24 and the document inclination / size correction unit 26 are instructed. However, the present invention is not limited to such a configuration. For example, when the photo paper size is selected via the screen shown in the lower part of FIG. 2 or when it is input that the original is a photo original, the original is a photo original. Only in this case, processing by the document inclination / size detection unit 24 and the document inclination / size correction unit 26 may be performed. In addition, a configuration for discriminating the document type is provided in the preceding stage of the document tilt / size detection unit 24, and when it is determined that the input image (document) is a photographic document (printed photograph document or photographic paper photograph document), the document skew The processing by the size detection unit 24 and the document inclination / size correction unit 26 may be performed. As the document type discrimination process, for example, a method disclosed in Japanese Patent Laid-Open No. 2002-232708 can be used. The document type determination process will be described below.

(1) The minimum density value (minimum pixel value) and the maximum density value (maximum pixel value) in each block are calculated for each block including n × m pixels (for example, 7 × 15 pixels) including the target pixel.
(2) A maximum density difference that is a difference between the calculated minimum density value and maximum density value is calculated.
(3) For each block, a total density busyness (for example, a sum of values calculated in the main scanning direction and the sub-scanning direction), which is a sum of absolute values of density differences between adjacent pixels, is calculated.
(4) The calculated maximum density difference is compared with the maximum density difference threshold, and the calculated total density busyness is compared with the total density busyness threshold.
When the maximum density difference <the maximum density difference threshold and the total density busyness <the total density busyness threshold, it is determined that the target pixel belongs to the background / photographic paper photograph area.
In other cases, it is determined that the target pixel belongs to the character / halftone dot region.
(5) For the pixels determined to belong to the background / photographic paper photograph region, the maximum density difference is compared with the background / photographic paper photograph determination threshold.
When the maximum density difference <the threshold value for determining the background / photographic paper photograph, it is determined that the target pixel belongs to the background area.
In other cases, it is determined that the target pixel belongs to the photographic paper photograph area (continuous gradation area).
(6) For pixels determined to belong to the character / halftone area,
The total density busyness is compared with the value obtained by multiplying the maximum density difference by the threshold value for character / halftone dot determination.
If the total density busyness <the maximum density difference × the threshold for character / halftone dot determination, it is determined that the pixel of interest belongs to the character area.
In other cases, it is determined that the target pixel belongs to the halftone dot region.
(7) The number of pixels classified into each of the background area, the photographic paper photograph area, the character area, and the halftone dot area is counted for each area, and compared with a predetermined threshold value for each area. The overall type is determined.
For example, the ratio of the number of pixels separated into the character area, the ratio of the number of pixels separated into the halftone dot area, and the ratio of the number of pixels separated into the photographic paper photograph area with respect to the number of pixels of the entire document. If the ratio of the character area is 30% or more of the total number of pixels, it is determined that the original is a character document. If the ratio of the halftone area is 20% or more of the total number of pixels, It is determined that the document is a point document (printed photograph document), and if the ratio of the photographic paper photograph area is 10% or more of the total number of pixels, the document is determined to be a photographic paper photograph document. Further, the ratio of the number of pixels separated into the character area to the number of pixels of the entire document is equal to or greater than the threshold value for the character area, and the ratio of the number of pixels separated into the halftone area is equal to that for the halftone area. If it is equal to or greater than the threshold, it is determined that the document is a character / halftone original (character / printed photo original).

  Next, specific processing contents by the document inclination / size detection unit 24, the display control unit 25, and the document inclination / size correction unit 26 will be described. FIG. 3 is a block diagram showing a configuration of the document inclination / size detection unit 24. The document inclination / size detection unit 24 includes an edge detection unit 241, an angle calculation unit 242, a coordinate information conversion unit 243, a document region detection unit 244, a cut region determination unit 245, a region display parameter generation unit 246, and a cut region selection unit 247. And a correction parameter generation unit 248 and the like. The operation of each unit provided in the document inclination / size detection unit 24 is controlled by the CPU or DSP provided in the image forming apparatus 100.

In the document inclination / size detection unit 24, RGB signals (hereinafter referred to as input image data) output from the input tone correction unit 23 are input to the edge detection unit 241.
Note that the document read by the color image input device 1 is not necessarily properly placed at a predetermined position on the document placement table, and may be placed at an angle. Accordingly, when the execution of the document inclination / size correction function is instructed via the setting screen shown in FIG. 2, the color image input apparatus 1 reads the document at the maximum readable size (for example, A3 size). . Therefore, when the execution of the document inclination / size correction function is instructed, the image inclination / size detection unit 24 receives image data read from the document at the maximum size. Note that the maximum size that can be read is a size of an area that includes a document and is wider than the document.

  The edge detection unit 241 detects the edge of the input image based on the input image data. Specifically, the edge detection unit 241 performs a reduction process on the input image data so as to obtain a predetermined resolution. The predetermined resolution is the optimum resolution for the edge detection process, and is set in advance and stored in a storage unit (not shown). In addition, the edge detection unit 241 extracts, as an edge, a pixel position where the difference between the pixel values of adjacent pixels is greater than a predetermined threshold for each line of the image data after the reduction process. Here, the predetermined threshold value is a value that can detect the boundary between the background of the document and the document region. For example, 3 is preset and stored in the storage unit.

  In the extracted edge (edge image), the edge detection unit 241 uses the leftmost pixel (edge) coordinates as the leftmost edge coordinates when scanning in the x-axis direction (lateral direction with respect to the document), and the rightmost pixel. The coordinates of (edge) are calculated as the right edge coordinates. The edge detector 241 stores the calculated coordinate values of the left edge coordinates and the coordinate values of the right edge coordinates in a coordinate information table as shown in FIG. FIG. 4 is a schematic diagram showing a coordinate information table. The coordinate information table is stored in, for example, a storage unit. The coordinate information table shown in FIG. 4 stores M coordinate values of left edge coordinates and right edge coordinates, and the number of coordinates M depends on the vertical size of the image data after reduction processing. .

Similarly, the edge detection unit 241 uses the coordinates of the uppermost pixel (edge) when scanned in the y-axis direction (vertical direction with respect to the document) as the upper edge coordinates and the coordinates of the lowermost pixel (edge). The lower edge coordinates are calculated, and the calculated upper edge coordinate values and lower edge coordinate values are stored in the coordinate information table. In the coordinate information table shown in FIG. 4, N coordinate values of the upper edge coordinate and the lower edge coordinate are stored, respectively, and the number of coordinates N depends on the size of the image data after the reduction process.
In the coordinate information table shown in FIG. 4, in addition to the left edge coordinates, right edge coordinates, upper edge coordinates, and lower edge coordinates, the left edge correction coordinates, right edge correction coordinates, upper edge Each coordinate value of correction coordinates and lower edge correction coordinates is stored. Since the coordinate values of the left edge correction coordinate, the right edge correction coordinate, the upper edge correction coordinate, and the lower edge correction coordinate are values calculated and stored by the coordinate information conversion unit 243 described later, the edge detection unit 241 At the end of the process according to, only the coordinate values of the left edge coordinates, right edge coordinates, upper edge coordinates and lower edge coordinates are stored in the coordinate information table.

  The angle calculation unit 242 calculates the inclination (angle) of the document based on the coordinate values of the edge coordinates (left edge coordinate, right edge coordinate, upper edge coordinate, and lower edge coordinate) detected by the edge detection unit 241. To do. As a method for calculating the inclination of the document, for example, an inclination detection method disclosed in Japanese Patent Application Laid-Open No. 11-331547 can be used.

  For each edge coordinate (left edge coordinate, right edge coordinate, upper edge coordinate or lower edge coordinate), the angle calculation unit 242 detects a change in inclination between the coordinates based on each coordinate value of each edge coordinate. Then, a point at which the inclination changes (edge change point) is detected, and a line segment connecting the detected edge change points is extracted. The angle calculation unit 242 selects, from the extracted line segments, the longest line segment of 45 degrees or less with respect to the x-axis direction (the horizontal direction with respect to the document, the sub-scanning direction), and the selected line segment The start point coordinates (StartEdgeX, StartEdgeY) and end point coordinates (EndEdgeX, EndEdgeY) of the minute are calculated. Then, the angle calculation unit 242 calculates tanα based on tanα = (EndEdgeY−StartEdgeY) / (EndEdgeX−StartEdgeX) from the start point coordinates and the end point coordinates, and refers to the tangent angle correspondence table shown in FIG. The inclination angle α is calculated. FIG. 5 is a schematic diagram showing a tangent angle correspondence table. The tangent angle correspondence table is stored in advance in a storage unit, for example.

The angle calculation unit 242 performs the above-described calculation for each edge coordinate (left edge coordinate, right edge coordinate, upper edge coordinate or lower edge coordinate), and with a line segment of 45 ° or less with respect to the x-axis direction. Select the longest line segment.
However, the line segment extracted based on the coordinate values of the left edge coordinate or the right edge coordinate is often inclined by 45 ° or more with respect to the x-axis direction. The line segment extracted based on the coordinate value is often 45 ° or less with respect to the x-axis direction. The original is often rectangular, and the line segment extracted based on the coordinate values of the upper edge coordinates and the line segment extracted based on the coordinate values of the lower edge coordinates have the same inclination. There are many. Therefore, the angle calculation unit 242 performs the above-described calculation for only one of the upper edge coordinates and the lower edge coordinates, selects the longest line segment of 45 ° or less with respect to the x-axis direction, The document inclination may be calculated based on the end point coordinates.

The coordinate information conversion unit 243 applies the coordinate values (EdgeX, EdgeY) of all the edge coordinates (left edge coordinates, right edge coordinates, upper edge coordinates, and lower edge coordinates) detected by the edge detection unit 241. Then, correction processing based on the angle α calculated by the angle calculation unit 242 is performed. The coordinate information conversion unit 243 performs correction processing based on the following formula, and calculates corrected coordinate values (CorrEdgeX, CorrEdgeY) obtained by correcting each coordinate value of each edge coordinate.

The coordinate values of each edge coordinate are stored in the coordinate information table as shown in FIG. 4, and the coordinate information conversion unit 243 converts the calculated corrected coordinate values (CorrEdgeX, CorrEdgeY) to the coordinate values before correction. Store them in the coordinate information table in association with each other.

The document area detection unit 244 extracts a valid document edge from the corrected coordinate values (CorrEdgeX, CorrEdgeY) of each edge coordinate calculated by the coordinate information conversion unit 243. The document area detection unit 244 calculates the coordinate value of the vertex of the corrected document area based on the corrected coordinate value extracted as a valid document edge. The coordinate values of the vertices of the corrected document area are the minimum x coordinate value (CorrEdgeMinX), the maximum x coordinate value (CorrEdgeMaxX), and the y coordinate value among the corrected coordinate values extracted as valid document edges. The minimum value (CorrEdgeMinY) and the maximum y-coordinate value (CorrEdgeMaxY). That is, the document area is an area surrounded by four vertices (CorrEdgeMinX, CorrEdgeMinY), (CorrEdgeMaxX, CorrEdgeMinY), (CorrEdgeMinX, CorrEdgeMaxY), (CorrEdgeMaxX, CorrEdgeMaxY).
The document area detection unit 244 compares the document inclination angle α calculated by the angle calculation unit 242 with the processing switching threshold when extracting the left edge and the right edge of the document, and determines the comparison result. The process is switched accordingly. The process switching threshold is set in advance to an optimum angle, and is a value of about 3 °.

6 and 7 are flowcharts showing a procedure of processing performed by the document area detection unit 244. After the coordinate information table as shown in FIG. 4 is generated by the processing by the edge detection unit 241 and the coordinate information conversion unit 243, the document region detection unit 244 performs the following document region detection processing.
The document area detection unit 244 calculates an effective count value for each coordinate value (each pixel) of the upper edge correction coordinates and the lower edge correction coordinates stored in the coordinate information table (S1). Specifically, the document area detection unit 244 determines that a difference between y coordinate values of correction coordinates existing at a plurality of different intervals with respect to the correction coordinate values of the upper edge and the lower edge is a predetermined threshold (for example, two pixels). The number of cases where the number is as follows is counted, and this count value is set as an effective count value for each correction coordinate.

  The document area detection unit 244 determines, for each of the upper edge and the lower edge, correction coordinates (edges) having an effective count value larger than a preset effective count parameter (for example, 2) from the effective count value for each correction coordinate. ) And the largest effective count value are calculated (S2). In the following, the average value of the valid count values calculated for the upper edge is AveEdgeCountT, and the maximum value is MaxEdgeCountT. Further, the average value of the effective count values calculated for the lower edge is AveEdgeCountB, and the maximum value is MaxEdgeCountB.

  The document area detection unit 244 sets the correction coordinate (edge) having the smallest y coordinate value among the correction coordinates having an effective count value equal to or greater than the average value AveEdgeCountT in each correction coordinate (each pixel) of the upper edge. Set to edge. In addition, the document area detection unit 244 sets the effective count value for the correction coordinate (edge) set as the upper maximum effective edge in the reliability (T_Conf) of the upper edge (S3). The average value AveEdgeCountT is not a fixed value but a parameter that is calculated and dynamically set in step S2, and can take a value of 0 to 88.

  The document area detection unit 244 determines the correction coordinate (edge) having the largest y-coordinate value among the correction coordinates having an effective count value equal to or greater than the average value AveEdgeCountB at the lower edge maximum correction coordinate (each pixel). Set to edge. Further, the document area detection unit 244 sets the effective count value for the correction coordinate (edge) set as the lower maximum effective edge in the reliability (B_Conf) of the lower edge (S4). The average value AveEdgeCountB is not a fixed value but a parameter that is calculated and dynamically set in step S2, and can take a value of 0 to 88.

  Based on the y coordinate value of the correction coordinate set as the upper maximum effective edge in step S3, the document area detection unit 244 determines whether the correction coordinates of the upper edge stored in the coordinate information table are valid or invalid. And the upper effective edge is extracted (S5). Specifically, the document area detection unit 244 calculates a difference between the y coordinate value of the correction coordinate set to the upper maximum effective edge and the y coordinate value of each correction coordinate of the upper edge stored in the coordinate information table. Calculate each. If the calculated difference is equal to or less than a predetermined range (for example, 2 pixels), the document area detection unit 244 determines that the correction coordinate (edge) is present (effective) on the extension line of the upper maximum effective edge, Extract as a valid edge. On the other hand, when the calculated difference is not less than or equal to the predetermined range, the document area detection unit 244 determines that the correction coordinate (edge) does not exist (invalid) on the extension line of the upper maximum effective edge and does not extract it as an effective edge. .

  Similarly, the document area detection unit 244 determines whether the correction coordinates of the lower edge stored in the coordinate information table are valid or invalid, and extracts the lower valid edge (S6). . Specifically, the document area detection unit 244 calculates the difference between the y coordinate value of the correction coordinate set to the lower maximum effective edge and the y coordinate value of each correction coordinate of the lower edge. If the calculated difference is equal to or smaller than the predetermined range, the document area detection unit 244 determines that the corrected coordinate (edge) exists on the extension line of the lower maximum effective edge (is effective) and extracts it as an effective edge. . On the other hand, if the calculated difference is not less than or equal to the predetermined range, the document area detection unit 244 determines that the correction coordinate (edge) does not exist (invalid) on the extension line of the lower maximum effective edge, and does not extract it as an effective edge. .

  FIG. 8 is a schematic diagram for explaining processing for determining whether or not each edge is valid. The upper side of FIG. 8 shows a diagram for explaining whether or not the correction coordinate (edge) of the upper edge is valid, and that the edge E1 is set as the upper maximum valid edge. To do. In this case, if the difference between the y-coordinate value of the edge E2 and the y-coordinate value of the edge E1 is less than or equal to a predetermined range, the edge E2 is extracted as an effective edge, and if the difference is not less than or equal to the predetermined range, the edge E2 is invalid. Determined. Further, the lower side of FIG. 8 shows a diagram for explaining the determination process of whether or not the correction coordinate (edge) of the lower edge is valid, and the edge E3 is set as the lower maximum valid edge. It shall be. In this case, if the difference between the y-coordinate value of the edge E4 and the y-coordinate value of the edge E3 is less than or equal to a predetermined range, the edge E4 is extracted as an effective edge, and if the difference is not less than or equal to the predetermined range, the edge E4 is invalid. Determined.

  The document area detection unit 244 performs the same processing as step S1, and calculates an effective count value for each coordinate value (each pixel) of the left edge correction coordinates and the right edge correction coordinates stored in the coordinate information table ( S7). Here, the document area detection unit 244 determines that a difference between x coordinate values of correction coordinates existing at a plurality of different intervals with respect to the correction coordinate values of the left edge and the right edge is a predetermined threshold (for example, two pixels). The number of cases where the number is as follows is counted, and this count value is set as an effective count value for each correction coordinate.

  In addition, the document area detection unit 244 performs the same processing as in step S2, and for each of the left edge and the right edge, a valid count parameter (for example, 2) set in advance from the valid count value for each correction coordinate. The average value of the effective count values of the correction coordinates (edge) having an effective count value larger than () and the largest effective count value are calculated (S8). In the following, the average value of the effective count values calculated for the left edge is AveEdgeCountL, and the maximum value is MaxEdgeCountL. Further, the average value of the effective count values calculated for the right edge is AveEdgeCountR, and the maximum value is MaxEdgeCountR.

  Here, the document area detection unit 244 compares the document inclination angle α calculated by the angle calculation unit 242 with the process switching threshold value, and determines whether the inclination angle α is equal to or greater than the threshold value (S9). . When it is determined that the inclination angle α is equal to or greater than the threshold (S9: YES), the document area detection unit 244 performs the same process as in step S3. Specifically, the document area detection unit 244 has a correction coordinate (x coordinate value having the smallest x coordinate value among correction coordinates having an effective count value equal to or greater than the average value AveEdgeCountL at each correction coordinate (each pixel) of the left edge. Edge) is set to the left most effective edge. Further, the document area detection unit 244 sets the effective count value for the correction coordinate (edge) set to the left maximum effective edge in the reliability (L_Conf) of the left edge (S10). Note that the average value AveEdgeCountL is not a fixed value but a parameter that is calculated and dynamically set in step S8 and can take a value of 0 to 88.

  Further, the document area detection unit 244 performs the same processing as in step S4, and among the correction coordinates having an effective count value equal to or greater than the average value AveEdgeCountR at each correction coordinate (each pixel) of the right edge, the x coordinate value The correction coordinate (edge) having the largest value is set as the right maximum effective edge. Further, the document area detection unit 244 sets the effective count value for the correction coordinates (edge) set to the right maximum effective edge in the reliability (R_Conf) of the right edge (S11). The average value AveEdgeCountR is not a fixed value but a parameter that is calculated and dynamically set in step S8 and can take a value of 0 to 88.

On the other hand, when the inclination angle α is not greater than or equal to the threshold (S9: NO), the document area detection unit 244 determines the left maximum effective edge candidate and the right maximum effective based on the correction coordinates of the left edge and the right edge. Three edge candidates are calculated (S12).
First, the document area detection unit 244 performs the same processing as steps S10 and S11, and calculates the first candidate of the left maximum effective edge and the right maximum effective edge.

  Next, the document area detection unit 244 is located outside (on the left side) the end (left end) of the upper effective edge and the lower effective edge extracted in steps S5 and S6, and the average value calculated in step S8. Among correction coordinates (pixels) having an effective count value equal to or greater than AveEdgeCountL, the outermost (left side) correction coordinates are specified. In other words, the document area detection unit 244 specifies the minimum x coordinate value among the x coordinate values of the corrected coordinates extracted as the upper effective edge and the lower effective edge, and sets an x coordinate value smaller than the specified x coordinate value. A correction coordinate (pixel) having an effective count value equal to or greater than the average value AveEdgeCountL is specified, and a correction coordinate having a minimum x coordinate value is specified among the specified correction coordinates. The document area detection unit 244 sets the specified correction coordinates as the second candidate for the leftmost maximum effective edge.

  Similarly, the document area detection unit 244 is located outside (on the right side) the end (right end) of the upper and lower effective edges extracted in steps S5 and S6, and the average value calculated in step S8. Among correction coordinates (pixels) having an effective count value equal to or greater than AveEdgeCountR, the outermost (right) correction coordinates are specified. That is, the document area detection unit 244 specifies the maximum x coordinate value among the x coordinate values of the correction coordinates extracted as the upper effective edge and the lower effective edge, and sets an x coordinate value larger than the specified x coordinate value. The correction coordinate (pixel) having an effective count value equal to or greater than the average value AveEdgeCountR is specified, and the correction coordinate having the maximum x coordinate value is specified among the specified correction coordinates. The document area detection unit 244 sets the specified correction coordinate as the second candidate for the rightmost maximum effective edge.

  Finally, the document area detection unit 244 is within a predetermined range centered on the ends (left end) of the upper effective edge and the lower effective edge extracted in steps S5 and S6, and the average value calculated in step S8. Among correction coordinates (pixels) having an effective count value equal to or greater than AveEdgeCountL, the outermost (left side) correction coordinates are specified. That is, the document area detection unit 244 specifies the minimum x coordinate value among the x coordinate values of the correction coordinates extracted as the upper effective edge and the lower effective edge, and falls within a predetermined range centered on the specified x coordinate value. Correction coordinates (pixels) having an x-coordinate value and an effective count value equal to or greater than the average value AveEdgeCountL are identified, and among the identified correction coordinates, the correction coordinate having the smallest x-coordinate value is identified. The document area detection unit 244 sets the specified correction coordinates as the third candidate for the leftmost maximum effective edge.

  Similarly, the document area detection unit 244 is within a predetermined range centered on the end (right end) of the upper effective edge and the lower effective edge extracted in steps S5 and S6, and the average calculated in step S8. Among the correction coordinates (pixels) having an effective count value equal to or greater than the value AveEdgeCountR, the outermost (right side) correction coordinates are specified. In other words, the document area detection unit 244 specifies the maximum x coordinate value among the x coordinate values of the correction coordinates extracted as the upper effective edge and the lower effective edge, and falls within a predetermined range centered on the specified x coordinate value. Correction coordinates (pixels) having an x-coordinate value and an effective count value equal to or greater than the average value AveEdgeCountR are identified, and among the identified correction coordinates, the correction coordinate having the maximum x-coordinate value is identified. The document area detection unit 244 sets the specified correction coordinates as the third candidate for the rightmost maximum effective edge.

  FIG. 9 is a schematic diagram showing candidates for the left maximum effective edge and the right maximum effective edge. The upper part of FIG. 9 shows an example of three candidates for the left maximum effective edge, and the lower part of FIG. 9 shows an example of three candidates for the right maximum effective edge. Between the two broken lines in the figure, a predetermined range centering on the ends of the upper effective edge and the lower effective edge is shown.

  The document area detection unit 244 selects one of the three candidates for the left maximum effective edge, selects one of the three candidates for the right maximum effective edge, and selects the left maximum effective edge and the right maximum effective edge. Is determined (S13). Here, when the above-mentioned third candidate (edges E7 and E10 in FIG. 9) is calculated for each of the left maximum effective edge and the right maximum effective edge, the third candidate is determined to be the left maximum The effective edge or the right maximum effective edge is determined. The third candidate is also extracted by limiting the range using the correction coordinates of the upper edge and the lower edge. Therefore, the third candidate is the most reliable candidate. If such a candidate exists, this candidate ( Edge) is the maximum effective edge.

  When the third candidate is not calculated, the second candidate (edges E6 and E9 in FIG. 9) is determined as the left maximum effective edge or the right maximum effective edge. The second candidate is extracted so as to be as wide as possible in order to prevent the content in the document from being lost. If the third candidate does not exist, this candidate (edge) is maximized. Edge. When the second candidate is not calculated, the first candidate (edges E5 and E8 in FIG. 9) is determined as the left maximum effective edge or the right maximum effective edge. The first candidate is extracted so as not to cause missing content in the document even if the actual document edge cannot be extracted. Therefore, both the second candidate and the third candidate are included. Only when it does not exist, this candidate (edge) is set as the maximum effective edge.

  The document area detection unit 244 sets the effective count value for the correction coordinate (edge) set to the left maximum effective edge to the reliability (L_Conf) of the left edge, and the correction coordinate ( The effective count value for the edge) is set to the reliability (R_Conf) of the right edge (S14).

  Based on the x coordinate value of the correction coordinates (pixels) set as the left maximum effective edge in step S10 or S13, the document area detection unit 244 determines the correction coordinates of the left edge stored in the coordinate information table. It is determined whether it is valid or invalid, and the left effective edge is extracted (S15). Specifically, the document area detection unit 244 calculates the x coordinate value of the correction coordinate set to the left maximum effective edge and the x coordinate value of each correction coordinate of the left edge stored in the coordinate information table. Each difference is calculated. If the calculated difference is equal to or less than a predetermined range (for example, two pixels), the document area detection unit 244 determines that the correction coordinate (edge) is present (effective) on the extension line of the left maximum effective edge. , Extracted as a valid edge. On the other hand, if the calculated difference is not less than or equal to the predetermined range, the document area detection unit 244 determines that the correction coordinate (edge) does not exist (invalid) on the extension line of the left maximum effective edge and extracts it as an effective edge. do not do.

  Similarly, the document area detection unit 244 determines whether it is valid or invalid for each correction coordinate of the right edge stored in the coordinate information table, and extracts the right effective edge ( S16). Specifically, the document area detection unit 244 calculates the difference between the x coordinate value of the correction coordinate (pixel) set to the right maximum effective edge and the x coordinate value of each correction coordinate of the right edge. . If the calculated difference is equal to or smaller than the predetermined range, the document area detection unit 244 determines that the correction coordinate (edge) exists on the extension line of the right maximum effective edge (is effective) and extracts it as an effective edge. To do. On the other hand, if the calculated difference is not less than or equal to the predetermined range, the document area detection unit 244 determines that the correction coordinate (edge) does not exist (invalid) on the extension line of the right maximum effective edge and extracts it as an effective edge. do not do.

  The document area detection unit 244 remains the effective edge for each correction coordinate (pixel) extracted as the upper effective edge, the lower effective edge, the left effective edge, or the right effective edge in steps S5, S6, S15, and S16, respectively. Or whether it is discarded as isolated point noise (S17). Specifically, the document area detecting unit 244 is outside (upper or lower) of the upper effective edge and the lower effective edge extracted in steps S5 and S6 in the left effective edge extracted in step S15. If there is an edge (pixel) existing in the area, it is determined whether the edge remains an effective edge or is isolated point noise. More specifically, the document area detection unit 244 determines that the left effective edge includes an edge having a y coordinate value smaller than the y coordinate value of the upper effective edge, and the y effective coordinate value of the lower effective edge. If there is an edge having a larger y coordinate value, it is determined whether the edge remains an effective edge or is isolated point noise.

  As a determination method, for example, when the number of valid edges existing within a predetermined range (for example, 3 pixels) from each edge is equal to or less than a predetermined number (for example, 2 pixels), the edge is invalidated as isolated point noise. Also, for example, when the correction coordinates (each pixel) of each edge are separated from the correction coordinates of the upper effective edge and lower effective edge by a predetermined distance (for example, 3 pixels) or more, the edge is invalid as isolated point noise. And

  Similarly, the document area detection unit 244 exists outside (upper or lower) of the upper and lower effective edges extracted in steps S5 and S6 in the right effective edge extracted in step S16. If there is an edge (pixel) to be used, it is determined whether this edge remains as an effective edge or is isolated point noise. Further, the document area detection unit 244 exists in the upper effective edge extracted in step S5 outside the left effective edge and the right effective edge extracted in steps S15 and S16 (left side or right side). If there is an edge (pixel), it is determined whether this edge remains as an effective edge or is isolated point noise. Further, the document area detection unit 244 exists outside (on the left or right side) the left and right effective edges extracted in steps S15 and S16 among the lower effective edges extracted in step S6. If there is an edge (pixel), it is determined whether this edge remains as an effective edge or is isolated point noise.

  FIG. 10 is a schematic diagram for explaining a process for determining whether or not it is isolated point noise. FIG. 10 shows a case where the left effective edge includes an edge E11 having a y coordinate value smaller than the y coordinate value of the upper effective edge. In this case, it is determined whether or not the edge E11 is isolated point noise based on the number of effective edges existing within a predetermined range from the edge E11, the correction coordinate value of the edge E11, or the like.

  The document area detection unit 244 performs the correction based on the corrected coordinate values of the effective edges (upper effective edge, lower effective edge, left effective edge, and right effective edge) determined to be effective edges in step S17. The coordinate value of the vertex of the document area is calculated (S18). The document area detection unit 244 has a minimum x coordinate value (CorrEdgeMinX), a maximum x coordinate value (CorrEdgeMaxX), a minimum y coordinate value (CorrEdgeMinY), and a maximum y coordinate value among the correction coordinate values of each effective edge. (CorrEdgeMaxY) is specified, and each coordinate value of the vertex of the corrected document area is set.

Returning to FIG. 3, processing by the cut region determination unit 245 will be described.
The cut region determination unit 245 calculates the coordinate values of the four vertices of the corrected document region calculated by the document region detection unit 244, the reliability of each edge (T_Conf, B_Conf, L_Conf, R_Conf), and the output image specified in advance. Based on the size information (OutSizeX, OutSizeY), an area to be cut out from the input image is determined. The cut area determination unit 245 specifies one or two cut areas, and calculates one or two sets of coordinate values of four vertices for specifying each cut area. Note that the coordinate values of the four vertices for specifying the cropping area are the minimum value of the x coordinate value (CropAreaMinX1, CropAreaMinX2), the maximum value of the x coordinate value (CropAreaMaxX1, CropAreaMaxX2), and the minimum value of the y coordinate value (CropAreaMinY1, CropAreaMinY2) ) And the maximum value of the y-coordinate values (CropAreaMaxY1, CropAreaMaxY2). That is, the cut region is a region surrounded by four vertices (CropAreaMinX, CropAreaMinY), (CropAreaMaxX, CropAreaMinY), (CropAreaMinX, CropAreaMaxY), (CropAreaMaxX, CropAreaMaxY). In the following, the coordinate values of the four vertices of the cropping area are abbreviated as (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1) and (CropAreaMinX2, CropAreaMaxX2, CropAreaMinY2, CropAreaMaxY2).

FIG. 11 is a flowchart illustrating a processing procedure performed by the cut-out area determination unit 245. The cut area determination unit 245 first calculates the following two parameters CorrEdgeDiffX and CorrEdgeDiffY based on the coordinate values of the vertices of the corrected document area.
CorrEdgeDiffX = CorrEdgeMaxX−CorrEdgeMinX
CorrEdgeDiffY = CorrEdgeMaxY−CorrEdgeMinY

The cut area determination unit 245 has the reliability (T_Conf) of the upper effective edge and the reliability (B_Conf) of the lower effective edge calculated by the document area detection unit 244 both equal to or higher than the threshold value (TH) for high reliability determination. It is determined whether or not (S21). When the reliability (T_Conf, B_Conf) is equal to or higher than the high reliability determination threshold (TH) (S21: YES), the cutout area determination unit 245 determines the absolute value (| OutSizeY−CorrEdgeDiffY |) of (OutSizeY−CorrEdgeDiffY |) And the absolute value (| OutSizeX−CorrEdgeDiffY |) of (OutSizeX−CorrEdgeDiffY)
｜ OutSizeY−CorrEdgeDiffY | <| OutSizeX−CorrEdgeDiffY |
It is determined whether or not (S22).

By determining whether or not | OutSizeY−CorrEdgeDiffY | <| OutSizeX−CorrEdgeDiffY |, whether the vertical / horizontal size of the area detected by the document area detection unit 244 is switched compared to the actual output image size. Can be determined.
When | OutSizeY−CorrEdgeDiffY | <| OutSizeX−CorrEdgeDiffY | (S22: YES), the cut area determination unit 245 executes the first cut area determination process on the assumption that the vertical and horizontal sizes are not switched (S23). Details of the first cut-out area determination process will be described later.

  If | OutSizeY−CorrEdgeDiffY | <| OutSizeX−CorrEdgeDiffY | is not satisfied (S22: NO), it is assumed that the vertical and horizontal sizes are switched, and the cut region determination unit 245 executes the second cut region determination process (S24). Details of the second cut-out area determination process will be described later.

If at least one of the reliability (T_Conf) of the upper effective edge and the reliability (B_Conf) of the lower effective edge is less than the high reliability determination threshold (TH) in step S21 (S21: NO), a cut region determination unit Whether the reliability of the left effective edge (L_Conf) and the reliability of the right effective edge (R_Conf) calculated by the document area detection unit 244 are both equal to or higher than the threshold value for high reliability determination (TH) 245. Is determined (S25). When the reliability (L_Conf, R_Conf) is equal to or higher than the high reliability determination threshold (TH) (S25: YES), the cutout area determination unit 245 determines the absolute value (| OutSizeX−CorrEdgeDiffX |) of (OutSizeX−CorrEdgeDiffX |) And the absolute value (| OutSizeY−CorrEdgeDiffX |) of (OutSizeY−CorrEdgeDiffX)
｜ OutSizeX−CorrEdgeDiffX ｜ <｜ OutSizeY−CorrEdgeDiffX ｜
It is determined whether or not (S26).

By determining whether or not | OutSizeX−CorrEdgeDiffX | <| OutSizeY−CorrEdgeDiffX |, whether the vertical / horizontal size of the area detected by the document area detection unit 244 is switched compared to the actual output image size. Can be determined.
When | OutSizeX−CorrEdgeDiffX | <| OutSizeY−CorrEdgeDiffX | (S26: YES), the cut area determination unit 245 executes the third cut area determination process on the assumption that the vertical and horizontal sizes have not been switched (S27). Details of the third cut-out area determination process will be described later.

If | OutSizeX−CorrEdgeDiffX | <| OutSizeY−CorrEdgeDiffX | is not satisfied (S26: NO), it is assumed that the vertical and horizontal sizes have been switched, and the cut region determination unit 245 executes the fourth cut region determination process (S28). Details of the fourth cut area determination processing will be described later.
If at least one of the reliability of the left effective edge (L_Conf) and the reliability of the right effective edge (R_Conf) is less than the high reliability determination threshold (TH) in step S25 (S25: NO), the cut region The determination unit 245 executes a fifth cut area determination process (S29). Details of the fifth cut region determination process will be described later.

Below, the 1st-5th cut area determination process is demonstrated.
When the first cut region determination process is performed, the cut region determination unit 245 sets the reliability of the left effective edge (L_Conf) and the reliability of the right effective edge (R_Conf) as a high reliability determination threshold (TH). And a set of four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1) are calculated according to the comparison result. That is, one cut area is specified.

The cut region determination unit 245 first calculates the following four parameters MerginY1, MerginY2, MerginX1, and MerginX2.
MerginY1 = (OutSizeY−CorrEdgeDiffY) / 2
MerginY2 = (OutSizeY−CorrEdgeDiffY) − (OutSizeY−CorrEdgeDiffY) / 2
MerginX1 = (OutSizeX−CorrEdgeDiffX) / 2
MerginX2 = (OutSizeX−CorrEdgeDiffX) − (OutSizeX−CorrEdgeDiffX) / 2

The cut region determination unit 245 arranges the upper effective edge and the lower effective edge so that each is located at equal intervals in the y-axis direction from the center of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY1
CropAreaMaxY1 ＝ CorrEdgeMaxY ＋ MerginY2

Next, the cutout region determination unit 245 compares the reliability (L_Conf) of the left effective edge and the reliability (R_Conf) of the right effective edge with the high reliability determination threshold (TH), and determines the reliability (L_Conf, When both R_Conf) are equal to or higher than the high reliability determination threshold (TH), the left and right effective edges are arranged at equal intervals from the center of the output image size in the x-axis direction. To do. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

When only the reliability (L_Conf) is equal to or higher than the high reliability determination threshold (TH), the cut region determination unit 245 arranges the left effective edge so as to match the left end of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX
CropAreaMaxX1 ＝ CorrEdgeMinX ＋ OutSizeX

When only the reliability (R_Conf) is equal to or higher than the high reliability determination threshold (TH), the cut region determination unit 245 arranges the right effective edge so as to match the right end of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMaxX−OutSizeX
CropAreaMaxX1 ＝ CorrEdgeMaxX

When the reliability (L_Conf, R_Conf) is less than the high reliability determination threshold (TH), the left and right effective edges are positioned at equal intervals in the x-axis direction from the center of the output image size. Arrange so that Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

  FIG. 12 is a schematic diagram showing a cut-out area. By the first cut area determination process described above, a cut area as shown in FIG. 12 is extracted. Specifically, when both the reliability levels (L_Conf, R_Conf) are equal to or higher than the high reliability determination threshold value (TH), a cutout region that substantially matches the document image is extracted. Further, when only the reliability (L_Conf) is equal to or higher than the high reliability determination threshold (TH), a cutout region with the left edge matched with the left edge of the original image is extracted, and only the reliability (R_Conf) is reliable. When it is equal to or greater than the degree determination threshold value (TH), a cutout region in which the right end coincides with the right end of the document image is extracted. When both the reliability levels (L_Conf, R_Conf) are less than the high reliability determination threshold value (TH), a cutout region with the center aligned with the center of the document image is extracted.

Next, the second cut area determination process will be described.
Also in the case of performing the second cut region determination process, the cut region determination unit 245 sets the reliability of the left effective edge (L_Conf) and the reliability of the right effective edge (R_Conf) to the high reliability determination threshold (TH ) And a set of four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1) are calculated according to the comparison result. That is, one cut area is specified.

The cut region determination unit 245 first calculates the following four parameters MerginY1, MerginY2, MerginX1, and MerginX2.
MerginY1 = (OutSizeX−CorrEdgeDiffY) / 2
MerginY2 = (OutSizeX−CorrEdgeDiffY) − (OutSizeX−CorrEdgeDiffY) / 2
MerginX1 = (OutSizeY−CorrEdgeDiffX) / 2
MerginX2 = (OutSizeY−CorrEdgeDiffX) − (OutSizeY−CorrEdgeDiffX) / 2

The cut region determination unit 245 arranges the upper effective edge and the lower effective edge so that each is located at equal intervals in the y-axis direction from the center of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY1
CropAreaMaxY1 ＝ CorrEdgeMaxY ＋ MerginY2

Next, the cutout region determination unit 245 compares the reliability (L_Conf) of the left effective edge and the reliability (R_Conf) of the right effective edge with the high reliability determination threshold (TH), and determines the reliability (L_Conf, When both R_Conf) are equal to or higher than the high reliability determination threshold (TH), the left and right effective edges are arranged at equal intervals from the center of the output image size in the x-axis direction. To do. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

When only the reliability (L_Conf) is equal to or higher than the high reliability determination threshold (TH), the cut region determination unit 245 arranges the left effective edge so as to match the left end of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX
CropAreaMaxX1 ＝ CorrEdgeMinX ＋ OutSizeX

When only the reliability (R_Conf) is equal to or higher than the high reliability determination threshold (TH), the cut region determination unit 245 arranges the right effective edge so as to match the right end of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMaxX−OutSizeX
CropAreaMaxX1 ＝ CorrEdgeMaxX

When the reliability (L_Conf, R_Conf) is less than the high reliability determination threshold (TH), the left and right effective edges are positioned at equal intervals in the x-axis direction from the center of the output image size. Arrange so that Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

  When the above-described second cut area determination process is performed, a cut area as illustrated in FIG. 12 is extracted in the same manner as the first cut area determination process.

Next, the third cut area determination process will be described.
When performing the third cut region determination process, the cut region determination unit 245 compares the reliability of the upper effective edge (T_Conf) and the reliability of the lower effective edge (B_Conf) with the high reliability determination threshold (TH), respectively. Then, a set of four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1) is calculated according to the comparison result. That is, one cut area is specified.

The cut region determination unit 245 first calculates the following four parameters MerginY1, MerginY2, MerginX1, and MerginX2.
MerginY1 = (OutSizeY−CorrEdgeDiffY) / 2
MerginY2 = (OutSizeY−CorrEdgeDiffY) − (OutSizeY−CorrEdgeDiffY) / 2
MerginX1 = (OutSizeX−CorrEdgeDiffX) / 2
MerginX2 = (OutSizeX−CorrEdgeDiffX) − (OutSizeX−CorrEdgeDiffX) / 2

The cut region determination unit 245 arranges the left and right effective edges so that the left and right effective edges are equally spaced in the x-axis direction from the center of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

Next, the cut region determination unit 245 compares the reliability (T_Conf) of the upper effective edge and the reliability (B_Conf) of the lower effective edge with the threshold value for high reliability determination (TH), and only the reliability (T_Conf) is obtained. When it is equal to or higher than the high reliability determination threshold (TH), the upper effective edge is arranged so as to coincide with the upper end of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY
CropAreaMaxY1 ＝ CorrEdgeMinY ＋ OutSizeY

When only the reliability (B_Conf) is equal to or higher than the high reliability determination threshold (TH), the cutout region determination unit 245 arranges the lower effective edge so as to coincide with the lower end of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMaxY−OutSizeY
CropAreaMaxY1 ＝ CorrEdgeMaxY

When the reliability (T_Conf, B_Conf) is both less than the threshold value for high reliability determination (TH), the upper effective edge and the lower effective edge are positioned at equal intervals in the y-axis direction from the center of the output image size. Arrange so that. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY1
CropAreaMaxY1 ＝ CorrEdgeMaxY ＋ MerginY2

  FIG. 13 is a schematic diagram showing a cut-out area. By the third cut area determination process described above, a cut area as shown in FIG. 13 is extracted. Specifically, when only the reliability (L_Conf) is equal to or higher than the high reliability determination threshold value (TH), a clipping region with the upper end matched is extracted from the upper end of the document image, and only the reliability (R_Conf) is extracted. Is equal to or higher than the high reliability determination threshold value (TH), a cut-out area with the lower end matched with the lower end of the document image is extracted. When both the reliability levels (L_Conf, R_Conf) are less than the high reliability determination threshold value (TH), a cutout region with the center aligned with the center of the document image is extracted.

Next, a fourth cut area determination process will be described.
Even when the fourth cut region determination process is performed, the cut region determination unit 245 sets the reliability of the upper effective edge (T_Conf) and the reliability of the lower effective edge (B_Conf) as the high reliability determination threshold (TH), respectively. Compare and calculate a set of four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1) according to the comparison result. That is, one cut area is specified.

The cut region determination unit 245 first calculates the following four parameters MerginY1, MerginY2, MerginX1, and MerginX2.
MerginY1 = (OutSizeX−CorrEdgeDiffY) / 2
MerginY2 = (OutSizeX−CorrEdgeDiffY) − (OutSizeX−CorrEdgeDiffY) / 2
MerginX1 = (OutSizeY−CorrEdgeDiffX) / 2
MerginX2 = (OutSizeY−CorrEdgeDiffX) − (OutSizeY−CorrEdgeDiffX) / 2

The cut region determination unit 245 arranges the left and right effective edges so that the left and right effective edges are equally spaced in the x-axis direction from the center of the output image size. Specifically, the minimum value (CropAreaMinX1) and the maximum value (CropAreaMaxX1) of the x coordinate value of the cut region are calculated based on the following formula.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX1
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX2

Next, the cut region determination unit 245 compares the reliability (T_Conf) of the upper effective edge and the reliability (B_Conf) of the lower effective edge with the threshold value for high reliability determination (TH), and only the reliability (T_Conf) is obtained. When it is equal to or higher than the high reliability determination threshold (TH), the upper effective edge is arranged so as to coincide with the upper end of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY
CropAreaMaxY1 ＝ CorrEdgeMinY ＋ OutSizeY

When only the reliability (B_Conf) is equal to or higher than the high reliability determination threshold (TH), the cutout region determination unit 245 arranges the lower effective edge so as to coincide with the lower end of the output image size. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMaxY−OutSizeY
CropAreaMaxY1 ＝ CorrEdgeMaxY

When the reliability (T_Conf, B_Conf) is both less than the threshold value for high reliability determination (TH), the upper effective edge and the lower effective edge are positioned at equal intervals in the y-axis direction from the center of the output image size. Arrange as follows. Specifically, the minimum value (CropAreaMinY1) and the maximum value (CropAreaMaxY1) of the y coordinate value of the cut region are calculated based on the following formula.
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY1
CropAreaMaxY1 ＝ CorrEdgeMaxY ＋ MerginY2

  When the above-described fourth cut area determination process is performed, a cut area as shown in FIG. 13 is extracted as in the third cut area determination process.

Next, a fifth cut area determination process will be described. 14 and 15 are flowcharts showing the procedure of the fifth cut area determination process.
When performing the fifth cut region determination process, the cut region determination unit 245 includes the left effective edge reliability (L_Conf), the right effective edge reliability (R_Conf), the upper effective edge reliability (T_Conf), and The reliability of the lower effective edge (B_Conf) is compared with the threshold value for high reliability determination (TH), and two sets of four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1), (CropAreaMinX2, CropAreaMaxX2, respectively) according to the comparison result CropAreaMinY2, CropAreaMaxY2) is calculated. That is, two cutout areas are specified.

The cut area determination unit 245 first calculates the following eight parameters MerginX11, MerginX12, MerginY11, MerginY12, MerginX21, MerginX22, MerginY21, MerginY22.
MerginX11 = (OutSizeX−CorrEdgeDiffX) / 2
MerginX12 = (OutSizeX−CorrEdgeDiffX) − (OutSizeX−CorrEdgeDiffX) / 2
MerginY11 = (OutSizeY−CorrEdgeDiffY) / 2
MerginY12 = (OutSizeY−CorrEdgeDiffY) − (OutSizeY−CorrEdgeDiffY) / 2
MerginX21 = (OutSizeY−CorrEdgeDiffX) / 2
MerginX22 = (OutSizeY−CorrEdgeDiffX) − (OutSizeY−CorrEdgeDiffX) / 2
MerginY21 = (OutSizeX−CorrEdgeDiffY) / 2
MerginY22 = (OutSizeX−CorrEdgeDiffY) − (OutSizeX−CorrEdgeDiffY) / 2

The cut region determination unit 245 determines whether or not the reliability (T_Conf) of the upper effective edge is equal to or higher than the high reliability determination threshold (TH) (S41).
As shown in the flowchart of FIG. 11, when the fifth cut region determination process is executed, at least one of the reliability of the left effective edge (L_Conf) and the reliability of the right effective edge (R_Conf) is high. It is less than the reliability determination threshold (TH), and at least one of the reliability of the upper effective edge (T_Conf) and the reliability of the lower effective edge (B_Conf) is less than the high reliability determination threshold (TH). Therefore, when the reliability (T_Conf) of the upper effective edge is equal to or higher than the high reliability determination threshold (TH), the reliability (B_Conf) of the lower effective edge is less than the high reliability determination threshold (TH), and the left When the reliability (L_Conf) of the partial effective edge is equal to or higher than the high reliability determination threshold (TH), the reliability of the right effective edge (R_Conf) is less than the high reliability determination threshold (TH).

When the reliability (T_Conf) is equal to or higher than the high reliability determination threshold (TH) (S41: YES), the clipping region determination unit 245 determines that the reliability (L_Conf) of the left effective edge is the high reliability determination threshold ( TH) or more is determined (S42). When the reliability (L_Conf) is equal to or higher than the high reliability determination threshold (TH) (S42: YES), the cut region determination unit 245 calculates the first region based on the following formula (S43), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMinX
CropAreaMaxX1 ＝ CorrEdgeMinX ＋ OutSizeX
CropAreaMinY1 ＝ CorrEdgeMinY
CropAreaMaxY1 ＝ CorrEdgeMinY ＋ OutSizeY
CropAreaMinX2 ＝ CorrEdgeMinX
CropAreaMaxX2 ＝ CorrEdgeMinX ＋ OutSizeY
CropAreaMinY2 ＝ CorrEdgeMinY
CropAreaMaxY2 ＝ CorrEdgeMinY ＋ OutSizeX

In step S42, when the reliability (L_Conf) is less than the high reliability determination threshold (TH) (S42: NO), the clipping region determination unit 245 determines that the reliability of the right effective edge (R_Conf) is high. It is determined whether or not the threshold value for determination (TH) is exceeded (S44). When the reliability (R_Conf) is equal to or higher than the high reliability determination threshold (TH) (S44: YES), the cut region determination unit 245 calculates the second region based on the following formula (S45), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMaxX−OutSizeX
CropAreaMaxX1 ＝ CorrEdgeMaxX
CropAreaMinY1 ＝ CorrEdgeMinY
CropAreaMaxY1 ＝ CorrEdgeMinY ＋ OutSizeY
CropAreaMinX2 ＝ CorrEdgeMaxX−OutSizeY
CropAreaMaxX2 ＝ CorrEdgeMaxX
CropAreaMinY2 ＝ CorrEdgeMinY
CropAreaMaxY2 ＝ CorrEdgeMinY ＋ OutSizeX

If the reliability (R_Conf) is less than the high reliability determination threshold (TH) in step S44 (S44: NO), the cut region determination unit 245 calculates the third region based on the following equation (S46). ), The process is terminated.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX11
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX12
CropAreaMinY1 ＝ CorrEdgeMinY
CropAreaMaxY1 ＝ CorrEdgeMinY ＋ OutSizeY
CropAreaMinX2 ＝ CorrEdgeMinX−MerginX21
CropAreaMaxX2 ＝ CorrEdgeMaxX + MerginX22
CropAreaMinY2 ＝ CorrEdgeMinY
CropAreaMaxY2 ＝ CorrEdgeMinY ＋ OutSizeX

When the reliability (T_Conf) of the upper effective edge is less than the high reliability determination threshold (TH) in step S41 (S41: NO), the cut region determination unit 245 determines that the reliability (B_Conf) of the lower effective edge is It is determined whether or not the threshold value is higher than the high reliability determination threshold (TH) (S47). When the reliability (B_Conf) is equal to or higher than the high reliability determination threshold (TH) (S47: YES), the clipping region determination unit 245 determines that the reliability (L_Conf) of the left effective edge is the high reliability determination threshold ( TH) or more is determined (S48). When the reliability (L_Conf) is equal to or higher than the threshold (TH) for high reliability determination (S48: YES), the cut region determination unit 245 calculates the fourth region based on the following formula (S49), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMinX
CropAreaMaxX1 ＝ CorrEdgeMinX ＋ OutSizeX
CropAreaMinY1 ＝ CorrEdgeMaxY−OutSizeY
CropAreaMaxY1 ＝ CorrEdgeMaxY
CropAreaMinX2 ＝ CorrEdgeMinX
CropAreaMaxX2 ＝ CorrEdgeMinX ＋ OutSizeY
CropAreaMinY2 ＝ CorrEdgeMaxY−OutSizeX
CropAreaMaxY2 ＝ CorrEdgeMaxY

In step S48, when the reliability (L_Conf) is less than the high reliability determination threshold (TH) (S48: NO), the clipping region determination unit 245 determines that the reliability of the right effective edge (R_Conf) is high. It is determined whether or not the threshold value for determination (TH) is exceeded (S50). When the reliability (R_Conf) is equal to or higher than the high reliability determination threshold (TH) (S50: YES), the cut region determination unit 245 calculates the fifth region based on the following formula (S51), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMaxX−OutSizeX
CropAreaMaxX1 ＝ CorrEdgeMaxX
CropAreaMinY1 ＝ CorrEdgeMaxY−OutSizeY
CropAreaMaxY1 ＝ CorrEdgeMaxY
CropAreaMinX2 ＝ CorrEdgeMaxX−OutSizeY
CropAreaMaxX2 ＝ CorrEdgeMaxX
CropAreaMinY2 ＝ CorrEdgeMaxY−OutSizeX
CropAreaMaxY2 ＝ CorrEdgeMaxY

If the reliability (R_Conf) is less than the high reliability determination threshold (TH) in step S50 (S50: NO), the cut region determination unit 245 calculates the sixth region based on the following equation (S52). ), The process is terminated.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX11
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX12
CropAreaMinY1 ＝ CorrEdgeMaxY−OutSizeY
CropAreaMaxY1 ＝ CorrEdgeMaxY
CropAreaMinX2 ＝ CorrEdgeMinX−MerginX21
CropAreaMaxX2 ＝ CorrEdgeMaxX + MerginX22
CropAreaMinY2 ＝ CorrEdgeMaxY−OutSizeX
CropAreaMaxY2 ＝ CorrEdgeMaxY

In step S47, when the reliability (B_Conf) is less than the high reliability determination threshold (TH) (S47: NO), the clipping region determination unit 245 determines that the reliability of the left effective edge (L_Conf) is high. It is determined whether or not the threshold value for determination (TH) is exceeded (S53). When the reliability (L_Conf) is equal to or higher than the high reliability determination threshold (TH) (S53: YES), the cut region determination unit 245 calculates the seventh region based on the following formula (S54), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMinX
CropAreaMaxX1 ＝ CorrEdgeMinX ＋ OutSizeX
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY11
CropAreaMaxY1 ＝ CorrEdgeMaxY + MerginY12
CropAreaMinX2 ＝ CorrEdgeMinX
CropAreaMaxX2 ＝ CorrEdgeMinX ＋ OutSizeY
CropAreaMinY2 ＝ CorrEdgeMinY−MerginY21
CropAreaMaxY2 ＝ CorrEdgeMaxY + MerginY22

In step S53, when the reliability (L_Conf) is less than the high reliability determination threshold (TH) (S53: NO), the clipping region determination unit 245 determines that the reliability (R_Conf) of the right effective edge is high. It is determined whether or not the threshold value for determination (TH) is exceeded (S55). When the reliability (R_Conf) is equal to or higher than the high reliability determination threshold (TH) (S55: YES), the cut region determination unit 245 calculates the eighth region based on the following formula (S56), and performs the processing. finish.
CropAreaMinX1 ＝ CorrEdgeMaxX−OutSizeX
CropAreaMaxX1 ＝ CorrEdgeMaxX
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY11
CropAreaMaxY1 ＝ CorrEdgeMaxY + MerginY12
CropAreaMinX2 ＝ CorrEdgeMaxX−OutSizeY
CropAreaMaxX2 ＝ CorrEdgeMaxX
CropAreaMinY2 ＝ CorrEdgeMinY−MerginY21
CropAreaMaxY2 ＝ CorrEdgeMaxY + MerginY22

If the reliability (R_Conf) is less than the high reliability determination threshold (TH) in step S55 (S55: NO), the cut region determination unit 245 calculates the ninth region based on the following equation (S57). ), The process is terminated.
CropAreaMinX1 ＝ CorrEdgeMinX−MerginX11
CropAreaMaxX1 ＝ CorrEdgeMaxX + MerginX12
CropAreaMinY1 ＝ CorrEdgeMinY−MerginY11
CropAreaMaxY1 ＝ CorrEdgeMaxY + MerginY12
CropAreaMinX2 ＝ CorrEdgeMinX−MerginX21
CropAreaMaxX2 ＝ CorrEdgeMaxX + MerginX22
CropAreaMinY2 ＝ CorrEdgeMinY−MerginY21
CropAreaMaxY2 ＝ CorrEdgeMaxY + MerginY22

  FIG. 16 is a schematic diagram showing a cut-out area. By the fifth cut area determination process described above, a cut area as shown in FIG. 16 is extracted.

Returning to FIG. 3, processing after the processing by the area display parameter generation unit 246 will be described. The area display parameter generation unit 246 is based on the four vertex coordinates (CropAreaMinX1, CropAreaMaxX1, CropAreaMinY1, CropAreaMaxY1), (CropAreaMinX2, CropAreaMaxX2, CropAreaMinY2, CropAreaMaxY2) for specifying the cutout area calculated by the cutout area determination unit 245. In the output image size, the start coordinates (StartX, StartY), the height h (length in the y-axis direction), and the width w (length in the x-axis direction) of the cut area indicated by each of the four vertex coordinates are calculated. The area display parameter generation unit 246 calculates each value using the following formula. Further, α is the document inclination angle calculated by the angle calculation unit 242.


h = CropAreaMaxY−CropAreaMinY
w = CropAreaMaxX−CropAreaMinX

FIG. 17 is a schematic diagram for explaining processing by the area display parameter generation unit 246. The upper side of FIG. 17 shows a clipping area before processing by the area display parameter generation unit 246, that is, before conversion of the input image into the coordinate system. Here, each vertex (specified by the document area detection unit 244) CorrEdgeMinX, CorrEdgeMaxX, CorrEdgeMinY, CorrEdgeMaxY) indicates a cut area. The lower side of FIG. 17 shows a cutout area after processing by the area display parameter generation unit 246, that is, after conversion of the input image into the coordinate system. Here, the cutout of the input image after conversion into the coordinate system is shown. A state in which the area is superimposed on the original image (input image) is shown.
The area display parameter generation unit 246 moves the coordinates (CropAreaMinX, CropAreaMaxX, CropAreaMinY, CropAreaMaxY) of the cut area specified by the cut area determination unit 245 from the coordinate system shown on the upper side of FIG. 17 to the lower side of FIG. It converts into the coordinate system shown, and acquires the start coordinate (StartX, StartY), height h, and width w of the cut-out area after conversion. The area display parameter generation unit 246 superimposes the cut area after conversion on the input image (data of the output image size), and stores the overlapped data in the image memory 5.

  The display control unit 25 reads the data (the output image size data obtained by superimposing the cut regions) generated by the region display parameter generation unit 246 and stored in the image memory 5. Further, the display control unit 25 acquires RGB signals (image data) input from the input tone correction unit 23 to the document inclination / size detection unit 24. The display control unit 25 may acquire the image data via the document inclination / size detection unit 24, or may read the image data from the image memory 5 once stored in the image memory 5.

  The display control unit 25 performs processing such as thinning out pixels on RGB image data (image data to be processed) so that the entire image based on the image data can be displayed on the operation panel 4. As the pixel thinning process, an interpolation method such as a nearest neighbor method, a bilinear method, or a bicubic method can be used. The nearest neighbor method is a method in which a pixel value of an existing pixel closest to a pixel to be interpolated (interpolated pixel) or a pixel value of an existing pixel having a predetermined positional relationship with respect to the interpolated pixel is used as the pixel value of the interpolated pixel. is there. The bilinear method is a method in which the pixel values of four existing pixels surrounding the interpolation pixel are weighted with values proportional to the distance from the interpolation pixel, and the average value of the obtained values is used as the pixel value of the interpolation pixel. In the bicubic method, a value calculated based on pixel values of a total of 16 existing pixels obtained by adding 12 existing pixels surrounding them to 4 existing pixels surrounding the interpolation pixel is used as the pixel value of the interpolation pixel. Is the method.

The display control unit 25 adds a rectangular frame based on the output image size data obtained by superimposing the cut areas to the image data after the thinning process, and generates image data (display data) for checking the cut area. To do. When two cut areas are extracted by the cut area determination unit 245, the display control unit 25 generates two pieces of cut area confirmation image data.
The display control unit 25 performs gradation correction (gamma correction) on the generated cut region confirmation image data based on the display characteristics of the display unit of the operation panel 4 and then causes the operation panel 4 to display the image data.

FIG. 18 is a schematic diagram showing a cut area confirmation screen, and FIG. 19 is a flowchart showing a cut area confirmation and correction process. The display control unit 25 causes the operation panel 4 to display confirmation screens such as those displayed on the upper and lower left sides of FIG. 18 (S61). The upper side of FIG. 18 shows an example in which a cutout area that substantially matches the area of the original image is detected (extracted), and the lower left side of FIG. 18 shows a position slightly shifted from the original image. An example in which a cutout area is detected is shown. When the user confirms the cut area extracted by the image forming apparatus 100 (cut area determination unit 245) on the confirmation screen, determines whether or not the displayed cut area is appropriate, and determines that it is appropriate If the OK button is operated and it is determined that it is not appropriate, the NG button is operated.
The operation panel 4 displaying the cut area confirmation screen determines whether or not the user has operated the OK button (S62), and when the user has operated the OK button (see the upper diagram in FIG. 18). This is notified to the display control unit 25, and the display control unit 25 notifies the cut region selection unit 247 to that effect.

  When the user operates the OK button (S62: YES), the cut area selection unit 247 outputs cut area information to the correction parameter generation unit 248 (S63), and ends the process. The information on the cut area may be the start coordinates (StartX, StartY), height h, and width w of the cut area generated by the area display parameter generation unit 246, or calculated by the cut area determination unit 245. Further, it may be a four-vertex coordinate for specifying the cut-out area. In addition, when the four vertex coordinates calculated by the cut region determination unit 245 are output to the correction parameter generation unit 248, the correction parameter generation unit 248 uses the cut four-vertex coordinates indicated by the four vertex coordinates acquired. The start coordinates (StartX, StartY), height h, and width w of the area are calculated.

  On the other hand, as shown in the lower left diagram of FIG. 18, when the user operates the NG button (S62: NO), the operation panel 4 notifies the display control unit 25 to that effect. In response to such notification, the display control unit 25 causes the user to select an effective edge in order to correct the cut region calculated by the cut region determination unit 245 (four vertex coordinates for specifying the cut region). Is displayed on the operation panel 4 (S64). An example of the selection screen is shown on the lower right side of FIG. 18, and an image (original image) based on the coordinate values of the four vertices of the corrected original area calculated by the original area detecting unit 244 is displayed. The user selects (instructs) an edge that is determined to be appropriately detected for the document image by appropriately operating the upper button, the lower button, the left button, or the right button. For example, the lower right diagram in FIG. 18 shows a state in which it is instructed that the upper edge and the left edge of the document image are appropriate.

  When the user operates the OK button in the state shown in the lower right diagram of FIG. 18, the operation panel 4 receives the edge selected as appropriate by the user (S65). The operation panel 4 notifies the cut region determination unit 245 of the edge selected by the user (the edge determined to be appropriate by the user) via the display control unit 25. The cut region determination unit 245 performs the cut region determination process again using the notified edge as an edge having a reliability equal to or higher than the high reliability determination threshold (TH) (S66).

  Then, based on the information (four vertex coordinates for specifying the cutting area) calculated again by the cutting area determination unit 245, the area display parameter generation unit 246 extracts information on the cutting area (starting coordinates and height of the cutting area). And width) again. Further, the display control unit 25 regenerates the image data for confirming the cut region based on the information calculated again by the region display parameter generating unit 246, and displays a confirmation screen based on the image data for confirming the cut region. 4 is displayed again (S67).

The operation panel 4 determines whether or not the user has operated the OK button (S68), and if the user has operated the OK button (S68: YES), this is indicated via the display control unit 25 in the cut-out area. Notify the selection unit 247. Then, the cut region selection unit 247 outputs the cut region information to the correction parameter generation unit 248 (S63), and ends the process. In this case as well, the cut area information may be the start coordinates (StartX, StartY), the height h and the width w of the cut area generated by the area display parameter generation unit 246, or the cut area determination. The four vertex coordinates calculated by the unit 245 for specifying the cutout area may be used.
In this way, by performing the cut region determination process again based on the edge that the user has determined to be appropriate, the burden on the user in the cut region correction work can be reduced.

  On the other hand, when the user operates the NG button (S68: NO), the operation panel 4 notifies the display control unit 25 to that effect. In this case, it is assumed that the document is difficult to automatically extract the cut area, and the user manually finely adjusts the cut area (four vertex coordinates for specifying the cut area). Accordingly, the display control unit 25 causes the operation panel 4 to display a setting screen for the user to finely adjust each edge of the cut area, and accepts designation of the cut area via the setting screen (S69).

  FIG. 20 is a schematic diagram showing a setting screen. The setting screen displays, for example, a document image and a frame indicating the cut area extracted by the cut area determination unit 245 as in the confirmation screen illustrated in FIG. The setting screen displays a + button and a − button corresponding to each of the upper edge, the lower edge, the left edge, and the right edge of the cut area. When the user operates the + button via the setting screen, the display control unit 25 displays the corresponding edge if the corresponding edge is the upper edge or the lower edge in the frame indicating the cut region displayed on the operation panel 4. If the corresponding edge is the left edge or the right edge, the corresponding edge is moved to the right. Further, when the user operates the − button via the setting screen, the display control unit 25 responds if the corresponding edge is an upper edge or a lower edge in the frame indicating the cut region displayed on the operation panel 4. If the corresponding edge is a left edge or a right edge, the corresponding edge is moved leftward.

  The user operates the OK button after moving each edge (each edge) of the cut region being displayed to a desired position. When the user operates the OK button via the setting screen, the display control unit 25 receives the cut area specified by the user via the operation panel 4. The display control unit 25 generates information on the cut region specified by the user (S70), outputs the generated cut region information to the correction parameter generation unit 248 (S63), and ends the process. The display control unit 25 changes the start coordinates (StartX, StartY), the height h, and the width w of the cut region generated by the region display parameter generation unit 246 according to the user's operation, and changes the changed information. What is necessary is just to output to the parameter generation part 248 for correction | amendment.

  Note that, for example, threshold values for confirmation determination (TH2, TH2> TH1) for determining whether or not to perform the cut region confirmation processing as described above are prepared, and each edge calculated by the document region detection unit 244 is prepared. When the reliability of all is equal to or greater than the threshold value for confirmation determination (TH2), the user may not perform confirmation via the confirmation screen. In this case, the cut region selection unit 247 outputs the information calculated by the region display parameter generation unit 246 to the correction parameter generation unit 248 without performing processing. Alternatively, the region display parameter generation unit 246 and the cut region selection unit 247 output the information calculated by the cut region determination unit 245 to the correction parameter generation unit 248 without performing processing.

Returning to FIG. 3, processing by the correction parameter generation unit 248 will be described. The correction parameter generation unit 248 includes the start coordinates (StartX, StartY), the height h and the width w of the cut region generated by the region display parameter generation unit 246, and the document inclination angle calculated by the angle calculation unit 242. α is converted into a value that matches the resolution of the image to be corrected by the document inclination / size correction unit 26 (same as the resolution of the input image), and is output to the document inclination / size correction unit 26 as a correction parameter.
The correction parameter generation unit 248 performs the same processing as the processing by the region display parameter generation unit 246 when the four-vertex coordinates for specifying the cutting region calculated by the cutting region determination unit 245 have been acquired. The start coordinates (StartX, StartY), height h, and width w of the cut area are calculated.

Next, correction processing by the document inclination / size correction unit 26 will be described. The document tilt / size correction unit 26 performs tilt / size correction processing on the RGB signals acquired from the input tone correction unit 23 based on the correction parameters acquired from the document tilt / size detection unit 24. When the input tone correction unit 23 temporarily stores the processed RGB signal in the image memory 5, the document inclination / size correction unit 26 reads the RGB signal from the image memory 5. As an inclination correction process (rotation process) performed by the document inclination / size correction unit 26, for example, an affine transformation process using the following rotation determinant is generally used. Note that (x ′, y ′) indicates a coordinate value after the coordinate (x, y) is rotated by an angle θ. Further, the trigonometric ratio (sin θ, cos θ) values used here can be calculated at high speed by using a table as shown in FIG.

Then, as the size correction process, the document inclination / size correction unit 26 performs an interpolation operation such as bilinear on the coordinates (x ′, y ′) after the rotation process, and calculates the pixel value of the output image after the size correction. To do. It should be noted that the interpolation pixel rotation processing necessary for calculating the pixel value of the coordinates (x ′, y ′) after the interpolation pixel rotation processing by the inverse transformation formula (the following determinant) of the rotation determinant described above. The previous coordinates (x _s , y _s ) are known. Therefore, the pixel value of the interpolation pixel is calculated based on the pixel value of the pixel that was in the vicinity of this interpolation pixel before the rotation process.

FIG. 21 is a schematic diagram for explaining the interpolation calculation. FIG. 21 shows an image before the rotation process, and the pixel values of the interpolation pixel Z (x _s , y _s ) are the pixel values (4) of the neighboring four pixels Z1, Z2, Z3, Z4 surrounding the interpolation pixel Z ( An example in which each pixel value is calculated based on Z1, Z2, Z3, and Z4) will be described. In FIG. 21, x _i ≦ x _s <x _{i + 1} , y _i ≦ y _s <y _{i + 1} , and the ratio of the distance between the interpolation pixel Z and the pixels Z1 and Z2 in the x coordinate is u: 1. -U, and the ratio of the distance between the interpolation pixel Z and the pixels Z1 and Z3 in the y coordinate is v: 1-v, and u and v are both numerical values of 0 or more and less than 1. The document inclination / size correction unit 26 calculates the pixel value (Z) of the interpolation pixel Z based on the following equation.

The coordinate value (x ′, y ′) of each pixel after the inclination / size correction processing by the original inclination / size correction unit 26 is 0 ≦ x ′ <corrected original image width, 0 ≦ y ′ <corrected This is the height of the original image. The document inclination / size correction unit 26 may perform the inclination correction process (rotation process) after performing the size correction process (interpolation calculation).
The document inclination / size correction unit 26 outputs the processed image data (RGB signal) to the region separation processing unit 27. The document skew / size correction unit 26 may store the processed image data in the image memory 5, and the region separation processing unit 27 may read the image data from the image memory 5.

  As described above, in the first embodiment, the edge of the document is specified by specifying the cutout area according to the reliability of each of the upper edge, the lower edge, the left edge, and the right edge extracted as the edges of the document. Even if the original is difficult to extract correctly, the detection accuracy of the cut-out area (original area) can be improved.

(Embodiment 2)
FIG. 22 is a block diagram illustrating a configuration of an image forming apparatus according to the second embodiment. The image forming apparatus 101 according to the second embodiment includes two front-stage color image processing apparatuses 2a and rear-stage color image processing apparatuses 2b instead of the color image processing apparatus 2 according to the first embodiment. The image forming apparatus 101 according to the second embodiment includes a large-capacity image storage device 6 such as a hard disk drive (HDD). The image forming apparatus 101 according to the second exemplary embodiment separates the image processing performed by the color image processing apparatus 2 according to the first exemplary embodiment into the first-stage image processing and the second-stage image processing, and the first-stage color image processing apparatus 2a and the second-stage image processing, respectively. This is a configuration executed by the color image processing apparatus 2b. In addition, about the structure similar to each part demonstrated in Embodiment 1, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  Specifically, the front color image processing apparatus 2a includes an A / D conversion unit 21, a shading correction unit 22, an input tone correction unit 23, a document inclination / size detection unit 24, and a display control unit 25, and a subsequent color image. The processing device 2b includes a document inclination / size correction unit 26, a region separation processing unit 27, a color correction unit 28, a black generation and under color removal unit 29, a spatial filter processing unit 30, an output gradation correction unit 31, and a gradation reproduction processing unit. 32.

  The pre-stage color image processing device 2 a stores the image data (RGB signal) processed by the input tone correction unit 23 in the image storage device 6. The pre-stage color image processing apparatus 2a also outputs the correction parameters calculated by the document inclination / size detection unit 24 to the image storage device 6 when the user instructs the execution of the document inclination / size correction function. The image data is stored in the image storage device 6 in association with the image data processed by the input tone correction unit 23.

On the other hand, the post-stage color image processing apparatus 2b reads out the image data and correction parameters stored in the image storage device 6, and executes the processes after the process by the document inclination / size correction unit 26. The document inclination / size correction unit 26 of the subsequent color image processing apparatus 2b sequentially stores the image data after the inclination / size correction processing in the image memory 5b, and the area separation processing unit 27 stores the image data in the image memory 5b. A region separation process is performed on the image data. Each processing after the region separation processing is performed by a pipeline method.
In the second embodiment, the same effect as in the first embodiment described above can be obtained. In the second embodiment, it is possible to asynchronously execute the processing by the preceding color image processing apparatus 2a and the processing by the subsequent color image processing apparatus 2b.

(Embodiment 3)
The image reading apparatus according to the present invention will be described below. FIG. 23 is a block diagram illustrating a configuration of an image reading apparatus according to the third embodiment. As shown in the figure, the image reading device 102 includes a color image input device 1 (image reading unit), a color image processing device 2, an operation panel 4, an image memory 5, an image storage device 6, and the like. The color image processing apparatus 2 includes an A / D conversion unit 21, a shading correction unit 22, an input tone correction unit 23, a document inclination / size detection unit 24, a display control unit 25, a document inclination / size correction unit 26, and a selector. 33 etc. In addition, about the structure similar to each part demonstrated in Embodiment 1, 2, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the color image processing apparatus 2 according to the third embodiment, the document inclination / size detection unit 24 supplies the image data (RGB signal) acquired from the input tone correction unit 23 together with the correction parameters to the document inclination / size correction unit 26. Output. The document inclination / size correction unit 26 performs processing on the image data acquired from the document inclination / size detection unit 24 based on the correction parameters acquired from the document inclination / size detection unit 24. The image data may be temporarily stored in the image memory 5 and then read by the document inclination / size correction unit 26.

  In the color image processing apparatus 2 according to the third embodiment, the image data processed by the input tone correction unit 23 and the image data processed by the document inclination / size correction unit 26 are input to the selector 33. . The selector 33 stores the image data input from the input tone correction unit 23 in the image storage device 6 when the user is not instructed to execute the document skew / size correction function. That is, image data that has not been processed by the document inclination / size correction unit 26 is stored in the image storage device 6. On the other hand, the selector 33 stores the image data input from the document inclination / size correction unit 26 in the image storage device 6 when the user instructs the execution of the document inclination / size correction function. That is, the image data subjected to the processing by the document inclination / size correction unit 26 is stored in the image storage device 6.

  When the execution of the document inclination / size correction function is instructed, the input tone correction unit 23 temporarily stores the processed image data in the image memory 5, and the document inclination / size correction unit 26 stores the image memory. The image data is read from 5, the inclination / size correction process is executed, and the processed image data is stored in the image memory 5 again. After the inclination / size correction processing for the image data of one page of the document is completed, the processed image data stored in the image memory 5 is read from the image memory 5 by the CPU or DSP of the image forming apparatus 100, and It is output to the image storage device 6. That is, in the third embodiment, the selector 33 switches whether the data transferred to the image storage device 6 is the image data after the input tone correction process or the image data after the inclination / size correction process. .

The image reading device 102 is not limited to the configuration in which the image data after the tilt / size correction processing is transferred to the image storage device 6, and for example, the image data is transmitted to a computer, a server, a digital multi-function peripheral, a printer, etc. The image reading apparatus 102 may be configured as described above.
In addition, the instruction whether or not to execute the tilt / size correction processing is not limited to the configuration performed via the operation panel 4. For example, in a computer, a scanner driver setting screen (setting of reading conditions of the image reading apparatus 102) is performed. Via a screen) using a mouse or a keyboard.

  The image forming apparatuses 100 and 101 described in the first and second embodiments are not limited to a digital color copying machine, but may be a digital color multifunction peripheral having a copier function, a printer function, a facsimile transmission / reception function, a scan to e-mail function, and the like. It may be applied. The digital color multi-function peripheral can further include a communication device such as a modem or a network card. In this case, for example, when transmitting a facsimile, if the modem performs a transmission procedure with the other party and secures the transmission possible state, the image data compressed in a predetermined format (the image read by the scanner) Data) is read from the memory, subjected to necessary processing such as changing the compression format, and sequentially transmitted to the other party via a communication line.

  Further, the image forming apparatuses 100 and 101 may be configured to perform data communication with a computer or other digital multi-function peripheral connected to the network via a network card, a LAN cable, or the like. Further, the present invention can be applied not only to a color multifunction peripheral but also to a multifunction peripheral that handles monochrome images.

  In the first to third embodiments, each unit including the document inclination / size detection unit 24 and the document inclination / size correction unit 26 provided in the image forming apparatuses 100 and 101 and the image reading apparatus 102 is realized by hardware logic. Alternatively, it may be realized by software using a processor such as a CPU or MPU. That is, the image forming apparatuses 100 and 101 and the image reading apparatus 102 include a CPU that executes instructions of a control program for realizing each function, a ROM that stores the control program, a RAM that develops various control programs, A storage device (recording medium) such as a memory for storing the control program and various data is provided.

  An object of the present invention is to provide a computer-readable recording medium on which a program code (execution format program, intermediate code program, source program) of a control program for causing a computer to execute each function is recorded. The program (CPU or MPU) is supplied, and the program code recorded in the recording medium is read and executed. In this case, a recording medium on which a computer program that realizes image processing (original tilt / size detection processing and tilt / size correction processing) according to the present invention is recorded can be provided in a portable manner.

  FIG. 24 is a schematic diagram showing a configuration example when the image processing apparatus according to the present invention is applied to a computer system. In a computer system 70 shown in FIG. 24, peripheral devices such as an image input device 75, an image display device 72, a keyboard 73, a mouse 74, and an image forming device 76 are connected to a personal computer 71. As the image input device 75, a flatbed scanner, a film scanner, a digital camera, a mobile phone, or the like is used. As the image display device 72, a CRT display, a liquid crystal display, or the like is used. As the image forming apparatus 76, an electrophotographic or inkjet printer is used.

  The personal computer 71 includes a reading device for reading a computer program recorded on the recording medium 77 according to the present invention. Therefore, by installing the computer program recorded on the recording medium 77 according to the present invention in the personal computer 71, various functions including the document inclination / size detection process and the inclination / size correction process described in the above-described embodiment are performed. It can be realized by the personal computer 71. Specifically, the personal computer 71 includes a CPU or MPU, and the CPU or MPU executes various processes by executing a predetermined computer program installed (loaded). The personal computer 71 may include a network card, a modem, or the like as communication means for connecting to a server or the like via a network.

  The recording medium 77 is an external storage device that can be attached to and detached from the personal computer 71, and is a program medium that can read a recorded program by being inserted into a reading device provided in the personal computer 71. There may be. Further, since the recorded program is processed by the microcomputer, the recording medium 77 may be a program medium such as a ROM.

  In any case, the stored program code may be configured to be accessed and executed by the microprocessor, or in any case, the program code is read and the read program code is the microcomputer. The program code may be downloaded to a program storage area (not shown) and executed. In this case, it is assumed that the download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, and includes a tape system such as a magnetic tape and a cassette tape, a magnetic disk such as a flexible disk and a hard disk, and a CD-ROM / MO / MD / DVD / Disk systems including optical disks such as CD-R, card systems such as IC cards (including memory cards) / optical cards, mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), A medium that carries a fixed computer program including a semiconductor memory such as a flash ROM can be used.

  Further, since the computer system 70 described above can be configured so that the personal computer 71 can be connected to a communication network including the Internet, the program code may be downloaded via the communication network. When the program code is downloaded from the communication network in this way, the program for downloading may be stored in the main device in advance or may be installed from another recording medium.

  The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite A communication network or the like is available. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  Each unit included in the image forming apparatuses 100 and 101 or the image reading apparatus 102 is not limited to a configuration realized by hardware logic or a configuration realized by software. For example, a part of each part of the image forming apparatuses 100 and 101 or the image reading apparatus 102 may be realized by hardware, and the remaining parts and hardware control may be realized by software.

  The preferred embodiments of the present invention have been specifically described above. However, each configuration, operation, and the like can be changed as appropriate, and are not limited to the above-described embodiments.

100 Image Forming Apparatus 1 Color Image Input Device (Image Reading Unit)
2 Color image processing device (image processing device)
3 Color image output device (image forming unit)
4 Operation panels (display unit, reception unit, change reception unit)
24 Document inclination / size detection unit 25 Display control unit (generation unit, extraction area change unit)
26 Document tilt / size correction unit 77 Recording medium 241 Edge detection unit 244 Document area detection unit (document edge extraction unit, reliability calculation unit, counting unit, calculation unit)
245 Cut region determination unit (region specifying unit, determination unit)

Claims (10)

In an image processing apparatus for extracting document image data read from the document from image data obtained by reading a wider area than the document including a rectangular document,
An edge detection unit for detecting an edge of an image based on the image data based on the image data;
A document edge extraction unit for extracting edges on each of the four sides of the document based on edges detected by the edge detection unit;
A reliability calculation unit for calculating the reliability of each edge extracted by the document edge extraction unit;
A reception unit for receiving the paper size of the output image;
And an area specifying unit that specifies an extraction area to be extracted as the document image data based on the reliability calculated by the reliability calculating unit and the paper size received by the receiving unit. apparatus. A determination unit that determines whether or not the reliability of each edge calculated by the reliability calculation unit is greater than or equal to a predetermined threshold;
When the determination unit determines that at least one edge is equal to or greater than a predetermined threshold, the area specifying unit extracts a paper size received by the reception unit based on an edge having a reliability greater than or equal to the predetermined threshold. The image processing apparatus according to claim 1, wherein an area is specified.   3. The image processing apparatus according to claim 1, further comprising: a generation unit that generates display data for displaying an image in which information indicating the extraction region specified by the region specifying unit is superimposed on an image based on the image data. An image processing apparatus according to 1. A display unit for displaying an image based on the display data generated by the generation unit;
A change accepting unit for accepting a change in the extraction area based on the image displayed by the display unit;
The image processing apparatus according to claim 3, further comprising: an extraction region changing unit that changes the extraction region specified by the region specifying unit based on a change received by the change receiving unit. The edge detection unit is configured to detect edges on the four sides of the image,
For each pixel constituting the edge detected by the edge detection unit, the number of pixels for which the difference value of the coordinate information between each pixel of interest and each pixel at a predetermined interval from the pixel of interest is less than a predetermined value is determined for each pixel of interest. A counting unit for counting as a count value;
A calculation unit that calculates an average value of effective count values counted by the counting unit for each edge,
The document edge extraction unit
Among the pixels constituting the edge on one side detected by the edge detection unit, the pixel having an effective count value larger than the average value calculated by the calculation unit, and the position farthest from the other side facing the one side Are extracted as edges on one side of the document,
Among the pixels constituting the edge on the other side detected by the edge detection unit, the pixel having an effective count value larger than the average value calculated by the calculation unit, and a pixel at a position farthest from the one side Extracted as the other side edge of the document,
Among the pixels constituting the edge on another side intersecting the one side detected by the edge detection unit, the pixel having an effective count value larger than the average value calculated by the calculation unit, and on the other side Extract a pixel at a position farthest from the other opposite side as an edge on the other side,
Among the pixels constituting the edge on the other other side detected by the edge detection unit, the pixel having an effective count value larger than the average value calculated by the calculation unit and farthest from the other side A pixel at a position is extracted as an edge on the other side, and
5. The reliability calculation unit according to claim 1, wherein the effective count value of the pixel extracted as each edge by the document edge extraction unit is used as the reliability of each edge. An image processing apparatus according to 1. An image processing apparatus according to any one of claims 1 to 5,
An image forming apparatus comprising: an image forming unit that forms an image based on image data processed by the image processing apparatus. An image reading unit for reading image data to obtain image data;
An image processing apparatus according to any one of claims 1 to 5,
An image reading apparatus, wherein an image read by the image reading unit is processed by the image processing apparatus. In an image processing method in which an image processing apparatus extracts a document image data read from the document from image data obtained by reading a region wider than the document including a rectangular document,
The image processing device detecting an edge of an image based on the image data based on the image data;
The image processing apparatus extracting each of the four side edges of the document based on the detected edges;
The image processing apparatus calculating a reliability of each extracted edge;
The image processing apparatus includes a step of specifying an extraction area to be extracted as the document image data based on the calculated reliability and the paper size of the output image received from the outside. . In a computer program for causing a computer to extract original image data read from an original from image data obtained by reading a region wider than the original including a rectangular original,
Causing the computer to detect an edge of an image based on the image data based on the image data;
Causing the computer to extract four side edges of the document based on the detected edges, respectively;
Causing the computer to calculate the reliability of each extracted edge;
And a step of causing the computer to specify an extraction area to be extracted as the document image data based on the calculated reliability and the paper size of the output image received from the outside.   A computer-readable recording medium on which the computer program according to claim 9 is recorded.