JP3955467B2 - Image processing program and image processing apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing program and an image processing apparatus, and more particularly, to an optical character reader (hereinafter referred to as OCR) that optically reads a surface image of a form and recognizes a predetermined area of the read image. The present invention relates to an image processing program and an image processing apparatus suitable for use.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, written characters entered on a form by handwriting or the like are optically read and recognized by an OCR and processed. In general, letters (referred to as pre-printing) that are unnecessary for reading a frame to be filled in, a yen in a monetary unit, or the like are printed on a form. Pre-printing of these frames and printed characters is not necessary for reading, and a color determined to be white by OCR, that is, a predetermined color called a dropout color is used. This dropout color is fixedly and strictly determined according to the wavelength of the light source or optical filter used in the photoelectric conversion unit. For this reason, the colors that can be used for pre-printing of forms are limited, and pre-printing is performed for forms printed in colors other than the specified color, such as color forms that have not been previously input by OCR. It was difficult to drop.
[0003]
As a conventional technique capable of dropping pre-printing on a form printed in a color other than the designated color, for example, a technique described in JP-A-6-176194 is known. This prior art is provided with a photoelectric conversion unit capable of outputting a color image, and based on the image of the reference form, by preparing in advance by color identification a part to be removed from the form image to be processed, The preprinted part of the form to be processed is dropped. With this conventional technique, it is possible to read the entered characters and recognize them by OCR without the restriction on the color used for pre-printing the form, but a reference form with no characters is prepared in advance. There is a need. In addition, in this prior art, when pre-printing such as a frame line and an entry character are in contact with each other and overlapping, no consideration is given to preventing the entry character from disappearing at the overlapping portion.
[0004]
Further, as a conventional technique related to image processing for causing an OCR to read, a technique described in JP-A-6-195509 is known. This prior art is provided with a photoelectric conversion unit that gives primary color components (red, green, blue) of each pixel from a multicolor image on the premise that the character color is black, and has the maximum value of these primary color components. A signal is output to remove the color and extract only the entered characters. However, the purpose of this prior art is to remove the color from the multicolored background of a document such as a check. If the character is colored, for example, if it is a blue character, the color is removed by removing the color. There is a problem that characters are also removed.
[0005]
Further, as another conventional technique, for example, a technique described in Japanese Patent Laid-Open No. 10-27213 is known. This conventional technology uses three primary colors of red, green, and blue from the photoelectric conversion unit so that characters written in chromatic colors, such as red pencils, red ballpoint pens, and blue ballpoint pens, can be read in the same way as achromatic characters. When a component is input, it is determined whether the color of the pixel belongs to an achromatic color, red color, green color, or blue color group, and the pixel is a character color specified in advance The maximum value of the three primary color components is output as a monochrome luminance signal of the pixel. If it is not a character color, the background level is output as a monochrome luminance signal for that pixel. And it converts into the monochrome luminance signal of the pixel. That is, multi-value image data is converted into binary image data. However, in this prior art, when preprints such as frame lines and written characters are in contact with each other and overlapped, consideration is given to leaving the written characters and dropping the preprints so that the overlapping portion does not disappear. Not.
[0006]
Further, as another conventional technique, a technique described in JP-A-6-290302 is known. This prior art performs provisional color separation on input image data when preprints such as frame lines and written characters come in contact with each other, and pixels that constitute each part of the provisional color separation The final color separation is determined by using the geometric information in combination with the density information of each portion and background. However, this prior art includes a binary image buffer, a gray image buffer, and a color image buffer. When a character targeted for a binary image is cut out and character recognition is performed, a specific portion of the input image, for example, a character When rechecking a portion where a hole is formed in a stroke or a contact portion between a ruled line and a character, the character recognition is performed by cutting out the character with reference to the above-described gray image buffer and color image buffer.
[0007]
For this reason, this conventional technique has a problem that it is necessary to access a gray image or a color image for each processing unit such as a character cutout unit or a character recognition unit, and the control becomes complicated. Further, this prior art does not generate and output a binary image in which the pre-printing is erased and the written characters are left, but the pre-printing is removed to correct the reading result, and the written characters are left. No consideration is given to displaying value images on the screen. In addition, this conventional technique uses linear geometric information such as growing an area that can be regarded as the same color. The reliability is not considered.
[0008]
[Problems to be solved by the invention]
The prior art described above has various problems as described together with the explanation of each prior art.
[0009]
An object of the present invention is to solve the above-mentioned problems of the prior art, and in a color form to be processed, a pre-print color such as a frame line is different from a pre-printed color for a form image. Image processing program and image capable of generating pre-printing and generating a binary image with the entry characters left, and performing highly accurate character recognition even if the print and the entry characters contact or overlap It is to provide a processing apparatus.
[0010]
[Means for Solving the Problems]
According to the present invention, the object is to input an input form image. Binarize the minimum value of the three primary color components Step and said A step of performing character recognition of the binarized image as a result of binarization, and a step of outputting the recognized character when the recognized character is accepted, and when the recognized character is rejected, Of the entered form image Pre-print color And The input color and Steps to set The form image No Drop out print color To generate a binarized image Step and said dropout And binarized binary image The step of character recognition and Output the recognized characters, and This is achieved by causing the information equipment to execute.
[0012]
Furthermore, the purpose is to input the form image. Binarize the minimum value of the three primary color components Means and said Means for performing character recognition of a binarized image obtained as a result of binarization; means for outputting a recognized character when the recognized character is accepted; and input when the recognized character is rejected. Of the form image Pre-print color And The input color and A means of setting The form image No Drop out print color Means for generating a binarized image by binarization And the dropout And binarized binary image Means for character recognition and , Means to output recognized characters and This is achieved by having
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of an image processing apparatus and a character reading system according to the present invention will be described in detail with reference to the drawings.
[0014]
FIG. 1 is a block diagram showing a configuration of a character reading system including an image processing apparatus according to an embodiment of the present invention. In FIG. 1, 100 is a color image scanner, 101 is an image processing unit, 102 is a color identification dropout binarization unit, 103 is a preprint / entry color identification unit, 104 is a preprint dropout binarization unit, Reference numeral 105 denotes a three primary color component minimum value binarization unit, 106 denotes a control unit, 107 denotes a character reading unit, 108 denotes an image storage unit, and 109 denotes a network.
[0015]
A character reading system according to an embodiment of the present invention includes a color image scanner 100 that collects a color image of a form surface, an image processing unit 104 that binarizes color image information from the color image scanner 100 into black and white, and 2 A character reading unit 107 such as an OCR for reading the digitized image data and an image storage unit 108 are connected by a network 109.
[0016]
The image processing unit 101 selects the minimum value of the three primary color components of the image data input from the color image scanner 100, that is, the data portion of the darkest part of the three primary color components, and performs the black and white binarization processing. A minimum value binarization unit 105, a color identification dropout binarization unit 102 that identifies preprinted colors and performs dropout processing on the colors, and a control unit 106 that controls selection and execution of these processes Composed. The three primary color component minimum value binarization unit 105 selects the minimum value of the three primary color components of the input color image and binarizes the primary color component as a gray value. As a method of binarizing the gray value, for example, as disclosed in Japanese Patent Application Laid-Open No. 11-120333, a method of determining a binarization threshold value from respective gray values of a pixel of interest and its surrounding pixels. It may be.
[0017]
The color identification dropout binarization unit 102 includes a color identification unit 103 that performs preprinting and color identification of written characters, and a binarization unit 104 that drops out a preprinted color using a color identification result. The character reading unit 107 performs character segmentation and character recognition from the binary image sent from the image processing unit 101, and the image storage unit 108 stores the binary image sent to the character reading unit 107. .
[0018]
FIG. 2 is a flowchart for explaining processing operations in the image processing unit 101 and the character reading unit 107 in the character reading system shown in FIG. 1, and this will be described next. Before the processing shown in FIG. 2 is started, the color image scanner 100 scans a form for character reading, collects a color image of the form, and transfers the image to the image processing unit 101. This image is temporarily stored and held in a buffer (not shown) provided in the color image scanner 100 or the image processing unit 101.
[0019]
(1) The binarization unit 105 for the minimum value of the three primary color components in the image processing unit 101 is the data of the minimum value of the three primary color components, that is, the darkest part of the three primary color components for the image from the color image scanner 100. A binary image obtained as a result of selecting a portion and performing the binarization process is transmitted to the network 109 via the control unit 106 (step 200).
[0020]
(2) The character reading unit 107 inputs the image binarized by the three primary color component minimum value binarizing unit 105 and designates an area to be read, that is, a field. The field can be specified by registering the field coordinates in advance as a form format (format) and selecting the field to select when reading the format, or by extracting the border of the form, A method of determining an area to be read from the arrangement and arrangement of frames may be used (steps 201 and 202).
[0021]
(3) Next, the character reading unit 107 reads characters in the field by cutting out characters in the field and executing character recognition. Next, it is determined whether the result of character reading in the field is accept or reject. As a result of this determination, if it is accepted, the read results are integrated and simply output (steps 203, 204, and 209).
[0022]
(4) If the result of character reading is rejected in the determination in step 204, the color image acquired by the color image scanner 100 is converted into the color of the preprinted / written character of the color identification dropout binarization unit 102. It inputs into the identification part 103 (step 205).
[0023]
(5) The color identification unit 103 performs color identification of preprinted portions such as form frame lines and guide character strings, and color identification of written characters from the input color image (step 206).
[0024]
(6) Next, the pre-print dropout binarization unit 104 uses the pre-print color from the color identification unit 103 to drop out the color and generate a binary image. If it is determined that pre-printing and written characters have the same color and the drop-out binarization is unsuccessful, the pre-print drop-out binarization unit 104 outputs a reject (step 207).
[0025]
(7) The character reading unit 107 reads characters in the field from the binary image transmitted from the pre-print dropout binarization unit 104 to the network 109, and integrates and outputs the read results ( Steps 208 and 209).
[0026]
As described above, the character reading system according to the embodiment of the present invention determines whether the result of the previous character reading in the field is accept or reject in the process of step 204, and in the case of reject, the pre-print is dropped out 2 Since character reading is performed on the value image, it is possible to prevent the pre-printed portion from becoming an obstacle to character recognition, and it is possible to improve the reliability of character reading.
[0027]
FIG. 3 is a block diagram showing the configuration of the color identification dropout binarization unit 102, FIG. 8 is a diagram for explaining a region for identifying the colors of preprints and written characters in the field, and FIG. 10 is a primary color component in FIG. 3 is a diagram illustrating a configuration of a selection unit 305. FIG. 3 and 10, reference numeral 301 denotes a pre-print color identification unit, 302 denotes an entry character color identification unit, 303 denotes a color component selection determination unit, 305 denotes a primary color component selection unit, 306 denotes a component value thickening / lightening unit, and 307. Is a binarization unit, 308 is a pixel identification unit, and the other symbols are the same as those in FIG.
[0028]
The color identification unit 103 includes a pre-print color identification unit 301, an entry character color identification unit 302, and a color component selection determination unit 303, and performs pre-printing and color identification of entry characters. Then, the blue component 310, the green component 311, and the red component 312 of the color image from the color image scanner 100 are input to the pre-print color identification unit 301 and the entry character color identification unit 302. The pre-print color identification unit 301 sets a process area for identification with respect to a partial color image in the field, and detects the color distribution of the pixels in the area to identify the pre-print color. . Note that the preprint color is not limited to one color, and may be a plurality of colors. For example, two colors of red and green may be extracted as the preprint colors.
[0029]
Next, with reference to FIG. 8, the area | region which identifies the color of the pre-print in a field and the entry character is demonstrated. In FIGS. 8A and 8B, the frame line 800 in the field 802 is colored, for example, a red or blue line. The entry character 801 is a reading target and is a black or colored character. The preprinted frame line 800 and the entry character 801 are shown as being entered in an overlapping state. When the color of the frame line is different from the color of the written characters, the frame line can be dropped out.
[0030]
In order to identify preprinted colors such as frame lines, color identification processing areas 803 and 804 are set as shown in FIG. In this processing region, pixel color identification is performed, the distribution of color pixels is obtained, and the preprint color is determined. Further, in order to identify the color of the entered character, a color identification processing area 813 is set as shown in FIG. In this processing area, when the pixel color is identified, the distribution of the color pixel is obtained, and the color of the written character is determined.
[0031]
Returning to the description of FIG. 3, the entry character color identification unit 302 sets a processing area for identification, for example, an area 813, for the partial color image in the field, and sets the color of the pixel in the area. The appearance distribution is detected and the color of the character is identified. Further, the color component selection determination unit 303 determines a primary color component to be selected as a binarized gray image among the three primary color components using the identified pre-print color and entry character color. For example, if the pre-print color is identified as red and the entry character color is identified as blue, the red component is determined as the primary color component to be selected. Further, if the pre-print color is identified as green and the entry character color is identified as black, the green component is determined as the primary color component to be selected. Then, the color component selection determination unit 303 outputs the preprint color to the primary color component selection unit 305 of the dropout binarization unit 104, and dropsout information 2 on the preprint color and entry character color. The value is output to the component value thickening / lightening unit 306 of the value converting unit 104.
[0032]
The above-described processing in the color identification unit 103 is performed on the color image of one field to identify the color. Then, after the above-described color identification is performed, the dropout binarization unit 104 described below uses the color image data in the same field to generate a monochrome binary image in which the pre-printed color is dropped out. Processing to generate is performed.
[0033]
As shown in FIG. 3, the dropout binarization unit 104 includes a primary color component selection unit 305, a component value thickening / lightening unit 306, a pixel color identification unit 308, and a grayscale binarization unit 307. In accordance with the result of the color component selection determination unit 303, the primary color component selection unit 305 simply selects one primary color component from among the three primary color components, that is, the blue component 310, the green component 311, and the red component 312 for each input pixel. Select and output the gray value.
[0034]
As shown in FIG. 10, the primary color component selection unit 305 selects one of the input blue component 310, green component 311, and red component 312 according to the color type information 1104 output from the color component selection determination unit 303. A primary color component is selected and output as a gray value 1105. Here, if the color type information 1104 is red, the red component 310 is output as the gray value 1105. If the color type information 1104 is green, the green component 311 is output as the gray value 1105. Furthermore, if the color type information 1104 is blue, the blue component 311 is output as a gray value 1105.
[0035]
In the above description, when the color type information 1104 output from the color component selection determination unit 303 indicates, for example, two colors of red and blue among the three primary color components as the pre-print color, the primary color component selection unit 305 may select the maximum value of the red component and blue component of each input pixel and output the maximum value as a gray value.
[0036]
The component value thickening / lightening unit 306 performs a process of reducing (darkening) or increasing (lightening) the magnitude of the lightness value of the input color. At this time, the pixel color identification unit 308 identifies the color of the pixel of interest, and sends the corresponding color type 322 to the darkening / lightening unit 306. The darkening / lightening unit 306 receives the pre-print color, the written character color, and the color of the pixel of interest, and darkens or lightens the gray value of the pixel of interest according to the type of the input color. For example, if the pre-print color is red, the entry character color is blue, and the pixel of interest is red, the pixel of interest is lightened. That is, the gray value of the pixel of interest is increased to approach white. The darkened / lightened gray value that is the output of the darkening / lightening unit 306 is input to the light / dark binarization unit 307 and is generated into a binary image. This density binarization unit 307 can be realized by, for example, a method disclosed in JP-A-11-120333.
[0037]
FIG. 4 is a diagram for explaining the processing operation for color identification of pre-printed / written characters in step 206 in the flow explained with reference to FIG. 2, which will be explained next. This process is executed separately by setting a region for each of the color identification of the pre-print and the color identification of the written characters.
[0038]
(1) First, the search range in the color image in the field is set. The search range is, for example, the processing areas 803 and 804 used for identifying the pre-print color and the processing area 813 used for identifying the entered character color as described with reference to FIG. 8 (step 400).
[0039]
(2) Next, the image in the area set in the field is scanned. The scanning procedure is performed by repeating the processing from step 403 to step 407 described below while scanning the target pixel in the horizontal direction and then in the vertical direction and scanning the image (step 401).
[0040]
(3) First, the color of the target pixel is determined while scanning the image. Here, as the color type of the target pixel, color identification is performed for four types of blue, green, red, and black, and the color type is obtained (step 403).
[0041]
(4) Next, the number of pixel colors determined in step 403 is counted. That is, if the target pixel is determined to be blue in step 403, the number of blue pixels is increased by one, and if the target pixel is determined to be green in step 403, the number of green pixels is increased by one. Similarly, when it is determined in step 403 that the target pixel is red or black, the number of red pixels and the number of black pixels are increased by one (steps 404 to 407).
[0042]
(5) Next, based on the counted number of each of the blue pixel, the green pixel, and the red pixel, the color type of pre-printing or written characters is determined. Here, among the number of blue pixels, green pixels, red pixels, and black pixels, the color having the maximum number is determined as the target color type, and the process ends (step 402).
[0043]
FIG. 5 is a diagram for explaining a character reading processing operation in the character reading unit 107, which will be described next.
[0044]
(1) First, a binary image in a set field is input, a character line that is a sequence of characters is extracted, and characters are cut out (steps 500 to 502).
(2) Character recognition of the cut out characters is performed, and knowledge processing is performed. The knowledge process is a process of preparing a dictionary of addresses and first and last names in advance and matching the character recognition result with the addresses and first and last names in the dictionary to improve the recognition accuracy (steps 503 and 504).
[0045]
FIG. 6 is a diagram for explaining an example of a color form to be processed. Next, this will be explained.
[0046]
For example, as illustrated in FIG. 6, frames 601, 603, and 608 are printed on the color form 600. In the illustrated example, characters 602 are entered in the frame 601. Also, pre-printed characters 604 and 605 are pre-printed on the frame 603, and characters 606 and 607 are entered. Further, a character 609 is entered in a frame 608. As in the example shown in the figure, there are forms in which the input characters and pre-printing are in contact with each other, and forms that overlap. For example, in the example shown in the figure, the entry character 602 and the frame 601 overlap. In addition, entry characters 606 and 607 and pre-prints 604 and 605 overlap. When pre-printing and written characters are in contact with each other or overlapped, it becomes an obstacle to character reading. Therefore, it is necessary to drop out pre-printing.
[0047]
Here, as an example, in a field where the frame 601 is red and the entry character 602 is blue, or in a field where the frame 603 and the preprints 604 and 605 are green, the frames 601 and 603 and the preprint characters 604 are preprinted. 605 is dropped out, that is, white, and the characters 602, 606, and 607 are black. Note that when the color of the frame 608 and the color of the entry character 609 are the same color, the pre-print dropout using the color information in the embodiment of the present invention is not applicable. As an example in which the color of the entry character 609 and the color of the pre-print frame 608 are the same color, the entry character 609 is green and the pre-print frame 608 is green, or the entry character 609 is black and the pre-print frame 608 is black. There are cases. In this way, it is difficult to perform pre-print dropout in the case of the same color, and in the case of the same color, both the pre-print and the input characters are output as black in the process of step 207 described with reference to FIG.
[0048]
FIG. 7 is a diagram illustrating an example of a binary image as a result obtained by performing image processing according to the embodiment of the present invention. The example shown in FIG. 7 shows a binary image 700 as a result of dropping out pre-printing from the color image of the color form 600 described with reference to FIG. In the illustrated binary image 700, the entry characters 602, 606, and 607 in FIG. 6 are processed as black as the entry characters 701, 702, and 703, and the preprints 601, 603, 604, and 605 are dropped out and become white. Yes. On the other hand, the pre-print 608 and the entry character 609 are not dropped out in the binary image 700 because of the same color, and both are expressed in black.
[0049]
FIG. 9 is a flowchart for explaining the details of the processing operation in step 403 of the pixel color determination described with reference to FIG. 4. Next, this will be described.
[0050]
(1) First, the three primary color components of the target pixel, that is, the red component value R, the green component value G, and the blue component value B are input. In general, the size of the color component of the pixel is expressed by 8-bit information, and each of R, G, and B takes a value of 0 to 225 (step 900).
[0051]
(2) Next, the red component value R is compared with other component values to determine whether or not the conditions of R> a · G and R> a · B are satisfied. If so, the pixel of interest is registered as red. In the above equation, a is a predetermined constant of 1 or more, and the same applies to the subsequent steps. When the constant a is set to a large value, only a pure color, in this case pure red, is determined as red (steps 901 and 902).
[0052]
(3) If the condition is not satisfied in the determination of step 901, the green component value G and the other component values are compared, and G> a · R and G> a · B are satisfied. It is determined whether or not the condition is satisfied. If the condition is satisfied, the target pixel is registered as green (steps 903 and 904).
[0053]
(4) If the condition is not satisfied in the determination in step 903, the blue component value B is compared with the other component values as described above, and B> a · R and B> a · It is determined whether or not the condition G is satisfied. If the condition is satisfied, the target pixel is registered as blue (steps 905 and 906).
[0054]
(5) Further, if the condition is not satisfied in the determination of step 905, then the red component value R, the green component value G, and the blue component value B are all smaller than a predetermined value c (R, G, B <c), and the maximum value of the red component value R, the green component value G, and the blue component value B is smaller than a value obtained by multiplying the minimum value by a constant d (Max (R, G, B) <d Whether or not Min (R, G, B)), that is, the three primary color component values are both small and the values are substantially the same, and the three primary color component values are both small and If the values are substantially the same, the pixel of interest is registered as black, and the process ends (steps 907 and 908).
[0055]
FIG. 11 is a block diagram illustrating the configuration of the component value thickening / lightening unit 306 in FIG. 3, and FIG. 12 illustrates an example of a thickening / lightening processing determination reference table used to determine the type of darkening / lightening processing. These will be described next. In FIG. 11, reference numeral 1200 denotes a thickening / lightening processing type determination unit, 1201 denotes a thickening multiplication unit, 1202 denotes a lightening multiplication unit, and 1203 denotes a selection unit.
[0056]
As shown in FIG. 11, the component value thickening / lightening unit 306 includes a thickening / lightening processing type determination unit 1200, a thickening multiplication unit 1201, a lightening multiplication unit 1202, and a selection unit 1203. It is prepared for. Then, the gray value 321 output from the primary color component selection unit 305, the entered character color and preprint color information 320 from the color component selection determination unit 303, and the color information 322 of the target pixel from the pixel color identification unit 308 are displayed. Are input to the component value thickening / lightening unit 306 shown in FIG.
[0057]
The thickening / lightening processing type determination unit 1200 uses the preprint color and entry color information 320 and the color information 322 of the target pixel to perform darkening output or lightening output of the input lightness value 321, or input It is determined with reference to the reference table 1300 for determining the darkening / lightening process which of the outputs to select the same value.
[0058]
The thickening / lightening process determination reference table used in the thickening / lightening process type determination unit 1200 is configured as shown in FIG. 12, and the horizontal item 1340 indicates the type of pre-print color, A case where the pre-print colors are red 1301, green 1302, blue 1303, and black 1304 is shown. Also, the vertical item 1342 indicates the type of entry character color, and indicates the case where the entry character color is red 1305, green 1306, blue 1307, and black 1308. Further, the vertical subdivision item 1341 indicates the type of the target pixel color, and the color of the target pixel may be red 1310, green 1311, blue 1312, or black 1313 for each entry character color.
[0059]
In the illustrated reference table 1300, contents indicating three types of darkening, lightening, and equivalence are stored and stored in advance in the case of each set of pre-print color, entry character color, and target pixel color. For example, if the pre-print color 1304 is green 1302 and the entry character color 1342 is red 1305, and if the target pixel color is red 1310, the darkening process indicated by 1320 is selected as the type of darkening / lightening process. If the target pixel color is green 1311, the lightening process indicated by 1321 is selected as the type of darkening / lightening process. If the target pixel color is blue 1312 or black 1313, the same value as the type of darkening / lightening processing, that is, no darkening or lightening is performed, and the value is output as it is.
[0060]
The thickening multiplication unit 1201 multiplies the input gray value 321 by a predetermined value to decrease the gray value, that is, outputs a dark density. Further, the lightening multiplication unit 1202 multiplies the input gray value 321 by a predetermined value, and increases the gray value, that is, outputs a light density. The selection unit 1203 selects either the density output or density output of the input density value 321 or the output of the same value as the input as 1213 according to the result output of the density / lightening processing type determination unit 1200. Output.
[0061]
In the embodiments of the present invention described above, the preprint color and entry character color of the form to be read are automatically identified and processed. However, the present invention is applied to the preprint color and entry character color. Can be included in the form identification number indicating the type of the read form, and the type of the read form can be set by reading the form identification number. In this case, the color identification dropout binarization unit 102 shown in FIG. 1 can be configured by the pre-print dropout binarization unit 104 alone.
[0062]
FIG. 13 is a flowchart for explaining processing operations in the image processing unit 101 and the character reading unit 107 in the character reading system according to another embodiment of the present invention when the form identification number is read and the type of the reading form is set. Next, this will be described. Note that this processing is performed after it is determined to be rejected in step 204 in the flow described with reference to FIG.
[0063]
(1) First, the type of the target form is set. Here, the type of form may be set by reading the form identification number described in the form, or the layout of the frame line may be extracted from the form image and collated with the layout of the prepared frame line. The form type may be set based on the form identification (step 1000).
[0064]
(2) Next, format information prepared in advance for each type of form set in step 1000, that is, format information is input. As the format information, information in which field position coordinates, pre-print colors, entry colors, etc. are prepared in advance for each form is used (step 1001).
[0065]
(3) Next, the processing from step 1003 to step 1008 described below is repeated until the field set as format information is exhausted, and the processing is terminated (step 1002).
[0066]
(4) The position coordinates of the field are set from the format information, and the pre-print color of the field is set from the format information. Further, the entry character color of the field is set from the format information (steps 1003 to 1005).
[0067]
(5) Next, a color image in the field is input, and binarization processing is performed to drop out the pre-print color using the pre-print color and the entry character color set in advance. Then, the characters in the field are read from the binarized image (steps 1006 to 1008).
[0068]
The above-described processing according to the embodiment of the present invention can be configured as a processing program, and this processing program is provided by being stored in a recording medium such as HD, DAT, FD, MO, DVD-ROM, or CD-ROM. be able to.
[0069]
As described above, according to the embodiment of the present invention, the means for identifying the color of the preprint / entry character from the color image in the field, and the means for generating the binarized image in which the identified preprint color is dropped out Since the character reading is retried with respect to the binarized image that has been dropped out, it is possible to improve the reliability of character reading in a field where pre-printing and written characters overlap. Further, it is possible to reduce the processing time by providing only a means for generating a binarized image in which the pre-printed color is dropped out.
[0070]
In addition, according to the embodiment of the present invention, since a means for setting the pre-print color and the entry character color of the field in the form is provided, even a form having a different pre-print color for each form or for each field. Pre-printing can be dropped out, and the reliability of character reading can be improved. Furthermore, based on the pre-print color, entry character color, and target pixel color, means for generating a grayscale image by selecting the maximum value of a single color component or multiple components from the three primary color components of the color image based on the pre-print color And a means for darkening or lightening the grayscale image, so that a gray value can be used for a single color dropout or a simultaneous dropout of a plurality of colors such as red and blue. Image quality can be improved.
[0071]
Further, according to the embodiment of the present invention, means for pre-registering a pre-print color or a text color as a form format for each field of the form image, and a pre-print color or entry of the registered field for each field Since a means for reading out the character color is provided, the image quality of the color dropout image can be improved with respect to a form whose format is known.
[0072]
Further, according to the embodiment of the present invention, means for setting a plurality of search areas in the field for each field of the form image, means for identifying the pre-print color in the predetermined search area, or another predetermined Since the means for identifying the character color of the entry within the search area is provided, it is possible to obtain an effect that color dropout can be performed even for a form whose format relating to the pre-print color is unknown.
[0073]
In addition, it will be as follows if the other embodiment of this invention demonstrated so far is described.
[0074]
1. In an image processing apparatus for generating a black and white binarized image for reading characters from a field of a form image, a binarized image generating means for binarizing a minimum value of three primary color components from a color image in the field, and a field An image processing apparatus comprising: means for generating a binarized image obtained by dropping out a preprint color from a color image therein.
[0075]
2. In an image processing apparatus for generating a black and white binarized image for reading characters from a field of a form image, a binarized image generating means for binarizing a minimum value of three primary color components from a color image in the field, and a field Means for generating a binarized image in which a pre-printed color is dropped out of a color image in the image, and means for generating a binarized image in which the pre-printed color is dropped out of the color image in the field is a form field. Means for setting a pre-print color or text color, means for selecting a maximum value of a single color component or a plurality of components from the three primary color components of the color image based on the pre-print color, and a pre-print Means for thickening or fading the grayscale image based on the color, the character color of interest and the pixel color of interest, and a method of binarizing the darkened or lightened grayscale image with a threshold value The image processing apparatus characterized by being composed of a.
[0076]
3. The means for setting the pre-print color or entry character color of the form includes means for pre-registering the pre-print color or entry character color as a form format for each field of the form image, and the field registered for each field. The image processing apparatus according to claim 2, comprising: a pre-print color or a character color for reading.
[0077]
4). Means for setting the preprint color or text color of the form, means for setting a plurality of search areas in the field for each field of the form image, means for identifying the preprint color in the predetermined search area, 3. The image processing apparatus according to claim 2, further comprising: means for identifying a character color for entry in another predetermined search area.
[0078]
5). In a character reading system including a character reading device that reads a black and white binarized image generated from a field of a form image and an image processing device that generates a black and white binarized image from the field of a form image, the image processing device includes: Binarized image generating means for binarizing the minimum value of the three primary color components from the color image in the field; means for generating a binarized image in which the pre-printed color is dropped out from the color image in the field; When the character reading device is caused to read a binarized image from a binarized image generating unit that binarizes the minimum value of the three primary color components, and the result is reject, the color image And means for retrying the reading of the binarized image from the means for generating the binarized image in which the pre-printed color is dropped out from Character reading system that.
[0079]
【The invention's effect】
As described above, according to the present invention, in a color form to be processed, pre-printing and entry characters are applied to a form image in which the color of pre-printing such as a frame line is different from the color of entry characters. Even if they are touched or overlapped, it is possible to erase the pre-printing and generate a binary image in which the entered characters remain, and to perform highly accurate character recognition.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a character reading system including an image processing apparatus according to an embodiment of the present invention.
FIG. 2 is a flowchart for explaining processing operations in an image processing unit and a character reading unit in the character reading system shown in FIG. 1;
FIG. 3 is a block diagram illustrating a configuration of a color identification dropout binarization unit.
FIG. 4 is a diagram for explaining the color printing processing operation for pre-printing / entry characters in step 206 in the flow described with reference to FIG. 2;
FIG. 5 is a diagram for explaining a character reading processing operation in a character reading unit;
FIG. 6 is a diagram illustrating an example of a color form that is a processing target.
FIG. 7 is a diagram illustrating an example of a binary image as a result obtained by performing image processing according to an embodiment of the present invention.
FIG. 8 is a diagram illustrating an area for identifying colors of pre-printing and entry characters in a field.
9 is a flowchart illustrating details of a processing operation in step 403 of pixel color determination described with reference to FIG. 4;
10 is a diagram illustrating a configuration of a primary color component selection unit in FIG. 3. FIG.
11 is a block diagram showing a configuration of a component value thickening / lightening unit in FIG. 3;
FIG. 12 is a diagram for explaining an example of a reference table for determining a thickening / lightening process used to determine a type of darkening / lightening process;
FIG. 13 is a flowchart illustrating a processing operation in a character reading system according to another embodiment of the present invention.
[Explanation of symbols]
100 color image scanner
101 Image processing unit
102 Color identification dropout binarization unit
103 Color identification part of pre-print / entry
104 Pre-print dropout binarization unit
105 3 primary color component minimum value binarization unit
106 Control unit
107 Character reader
108 Image storage unit
109 network
301 Pre-print color identification unit
302 Character color identification part
303 Color component selection determination unit
305 Primary color component selector
306 Concentration / lightening part of component value
307 binarization part
308 Pixel identification unit
1200 Thickening / lightening processing type decision unit
1201 Thickening multiplier
1202 Multiplier for lightening
1203 selection unit

Claims (7)

Binarizing the minimum value of the three primary color components of the input form image;
Performing character recognition of the binarized image resulting from the binarization;
Outputting the recognized character when the recognized character is accepted, and
If the recognized character is rejected, and setting a pre-printed color of the inputted document image and serial and incoming text color, and binarized by dropout flop Les printing color of the form image 2 A program that causes an information device to execute a step of generating a value image, a step of performing character recognition of the binarized image binarized by dropout, and a step of outputting the recognized character . 2. The program according to claim 1 , further causing the information device to execute a step of reducing (darkening) or increasing (lightening) the gray value of the input color . 3. The program according to claim 1, wherein the pre-print color and the entry character color are set for each field of the form image . The pre-print color or entry character color of the form is registered in advance as a form format for each field of the form image, and the pre-print color of the registered field for each field. The program according to claim 3 , wherein the program is executed by reading out the character color . Means for binarizing the minimum value of the three primary color components of the input form image;
Means for performing character recognition of the binarized image resulting from the binarization;
Means for outputting the recognized character when the recognized character is accepted;
If the recognized character is rejected, and means for setting the pre-print color of the inputted document image and serial and incoming text color, and binarized by dropout flop Les printing color of the form image 2 An image processing apparatus comprising: means for generating a value image; means for performing character recognition of the binarized image binarized by dropout; and means for outputting the recognized character . 6. The image processing apparatus according to claim 5 , further comprising a component value thickening / lightening means for reducing (darkening) or increasing (lightening) the gray value of the input color . 7. The image processing apparatus according to claim 5, wherein the pre-print color and the text color are set for each field of the form image .

Images