JP2004341656A - Image processor and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To select a binarization parameter suitable for an objective part. <P>SOLUTION: When the image data of a predetermined slip are inputted from an image scanner 1 to an image processor 2, an identification information reading means 3 reads slip identification information from the image data, and transmits it to a definition information retrieving means 5. The definition information retrieving means 5 retrieves a definition information storing part 4 based on the slip identification information, and extracts the definition information corresponding to a target slip, and stores it in a binarization information storing part 6, and transfers it to a threshold selecting means 7. The threshold selecting means 7 selects the optimal threshold by using image data in an item area for each of the items of the slip based on the definition information. Also, the binarization parameter is set based on the definition information. The threshold and the binarization parameter are stored in the binarization information storing part 6. A binarization processing means 8 executes the binarization processing of the image data in the image area for each item based on the binarization information. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing apparatus and an image processing program, and more particularly to an image processing apparatus and an image processing program for performing image processing for binarizing image information of a form optically read using a predetermined threshold value.
[0002]
[Prior art]
Conventionally, in order to improve the efficiency of form reading operations, form reading devices that read optical image information of a form placed on a table and process the image information to perform character recognition etc. have been provided to financial institutions, etc. Has been introduced.
[0003]
The form reading device includes a non-contact image scanner that captures a two-dimensional area including a form to obtain an entire image (image information), and an image processing device that performs image processing of the image information.
[0004]
The operation of the form reading device will be described. FIG. 15 is a diagram illustrating an example of a form. The figure is an example of a form (slip) used in a bank. The form 90 in the example of the figure includes guide letters 91 indicating a plurality of items such as “bank name”, “branch name”, “payee”, “client”, “account number”, “amount”, and the like. A frame 92 for entering data corresponding to the item and a seal column 93 for pressing a seal are printed in advance. For example, the customer fills in the corresponding numbers and characters in each frame 92 of the form 90 and seals the seal column 93. The entire image (image information) of the form 90 placed on the table by the operator is read by a non-contact type image scanner and transferred to the image processing apparatus.
[0005]
The image processing device binarizes the acquired image information using a predetermined threshold to generate image data (display data) of the entire form for display, and performs character recognition of the data in each frame 92. Do. The threshold value used for binarization is dynamically selected for each form according to the density level of the acquired image information of the entire form (for example, see Non-Patent Document 1).
[0006]
[Non-patent document 1]
Nobuyuki Otsu, "Automatic Threshold Selection Method Based on Discrimination and Least Square Criterion", IEICE Transactions, J63-D, vol. 4, 1980, p. 349-356
[0007]
[Problems to be solved by the invention]
However, in the conventional image processing apparatus, since the threshold value is selected from the density level of the entire form, there is a problem that the selected threshold value may not be suitable for a target location.
[0008]
As described above, when binarizing image information, a threshold value is selected from the density level of the entire form, but the form includes guide characters 91 and a frame 92 printed in advance, and the like. Information data having different density levels, such as handwritten characters such as numbers and characters entered in the frame 92, are mixed. For this reason, the threshold value suitable for the handwritten characters to be originally output is not affected by the density level of the guide character 91, the black print frame, or the dropout color removed at the time of character recognition in the form. Case occurs. If the threshold value does not match the density level of the handwritten character, for example, a thin portion of the character is not output during binarization, and the character after binarization is blurred, and character recognition is difficult. Problem arises. Further, in generating an image for display, a problem that a part to be converted into information such as a guide character may be lost may occur.
[0009]
In general, when binarizing areas for different uses, such as a recognition area for character recognition, a display area for display, and a seal area for acquiring seal information, binarization processing is performed in accordance with each area. It is desirable to make a fine adjustment of. However, conventionally, since uniform processing has been performed as a whole, adjustment according to the characteristics of each region has not been performed.
[0010]
As described above, by automatically selecting the threshold value based on the density level of the entire form, the threshold value is determined by the color composition of the form (headings and guide characters) and the ratio of the black form to the entire form. There was a problem that would be done. Also, it has been difficult to finely adjust the binarization processing according to the purpose of character recognition or display. As described above, the binarization parameters referred to at the time of binarization including the threshold value are not optimized for each area, and as a result, there arises a problem that handwritten characters are blurred or display information is lost. Was.
[0011]
The present invention has been made in view of such a point, and an object of the present invention is to provide an image processing apparatus and an image processing program capable of selecting a binarization parameter suitable for a target portion.
[0012]
[Means for Solving the Problems]
In the present invention, an image processing apparatus as shown in FIG. 1 is provided in order to solve the above problems. An image processing apparatus 2 according to the present invention is connected to an image scanner 1 that reads optical image information of a form, and performs image processing of image data generated by the image scanner 1.
[0013]
In the form to be read, in addition to the form identification information for identifying the form, each item information is described at a predetermined position. Definition body information that defines the position and range of each item is stored in the definition body information storage unit 4 for each form.
[0014]
In the image processing apparatus 2, when image data of a form read by the image scanner 1 is input, the identification information reading means 3 reads form identification information for identifying the form from the image data of the form. The definition body information search unit 5 searches the definition body information storage unit 4 based on the form identification information acquired by the identification information reading unit 3, and extracts the definition body information of the corresponding form. The extracted definition body information is transmitted to the threshold value selecting unit 7 by being stored in the binarized information storage unit 6 or the like. The threshold value selecting unit 7 extracts image data of the item area for each item based on the obtained definition body information, and selects a threshold value according to the density level of the image data of the item area. The binarization processing means 8 binarizes the item area for each item using a threshold value set for each item area, and generates binarized data.
[0015]
As described above, in the image processing apparatus 2 according to the present invention, the optimum threshold value is selected for each item based on the definition body information, and the binarization process is performed for each item region using the threshold value. Image data is output.
[0016]
Further, in order to solve the above-described problem, the computer stores, in a predetermined storage unit, definition body information including a position and a range of the item in the form defined for each item configuring the form in a predetermined storage unit, Reading form identification information for identifying the form, searching predetermined storage means storing the definition form information based on the form identification information, and extracting definition form information corresponding to the form; Selecting a threshold value according to the density level of the image information of the item area defined for each item based on body information; and, for each item, the image information of the item area using the threshold value. And v. Performing a value conversion process.
[0017]
By causing the computer to execute such an image processing program, the computer functions as the image processing device 2. The computer stores definition body information defining each item constituting the form in the definition body information storage unit 4, and when image data read by the image scanner 1 is input, a form for identifying the form from the image data. The identification information is read, and the definition body information of the corresponding form is extracted based on the form identification information. Then, a threshold value for binarization is selected based on the definition body information extracted for each item, and the image data of each item region is subjected to a binarization process using the selected threshold value.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. First, the concept of the invention applied to the embodiments will be described, and then the specific contents of the embodiments will be described.
[0019]
FIG. 1 is a conceptual diagram of the invention applied to the embodiment of the present invention.
An image processing apparatus 2 according to the present invention is connected to an image scanner 1 for reading optical image information of a form and outputting the image data, and includes identification information reading means 3 for reading form identification information from the image data; A definition body information storage unit 4 for storing the definition body information defining each item, a definition body information search unit 5 for searching for the corresponding definition body information based on the form identification information, and binarization including the definition body information Binarization information storage unit 6 for storing binarization information, threshold selection means 7 for selecting a threshold for binarization, and each item area based on the threshold and the binarization information And a binarization processing means 8 for performing the binarization processing.
[0020]
The image scanner 1 reads optical image information of a form placed on a table, generates image data thereof, and delivers the image data to the image processing apparatus 2.
The identification information reading means 3 reads form identification information for identifying the form, which is recorded at a predetermined position on the form, from the image data read by the image scanner 1. The read form identification information is sent to the definition body information search means 5.
[0021]
The definition body information storage unit 4 stores definition body information that defines each item constituting a form to be processed by the image processing apparatus 2. The definition body information includes, for example, coordinates indicating the position of each item constituting the form, a range such as a width and a height of an area, a use purpose of each item (for display / character recognition, and the like), and a type of a written character. Definitions such as (handwriting / printing / printing / printing) are informationized. The definition body information is prepared for each form.
[0022]
The definition body information search means 5 searches the definition body information stored in the definition body information storage section 4 using the form identification information read by the identification information reading means 3 and extracts the definition body information of the corresponding form. . The extracted definition body information of the target form is transferred to the threshold value selecting means 7 by storing it in the binarized information storage unit 6, for example.
[0023]
The binarization information storage unit 6 stores information necessary for executing binarization processing. The binarization information includes a threshold value used for the binarization processing and a binarization parameter for adjusting the binarization processing. The threshold value is a value selected for each area of each item according to the density level of the area. The binarization parameter is a parameter for adjusting the binarization processing according to a use such as display or character recognition. In addition, for example, the definition body information of the target form extracted by the definition body information searching means 5 is stored.
[0024]
The threshold value selecting means 7 sets a threshold value used in the binarization process and a binarization parameter based on the definition body information, and stores them in the binarization information storage unit 6. Is transferred to the binarization processing means 8. As the threshold value, for example, histogram information of image data of each item area defined by the definition information is created, and an optimal threshold value is selected based on the histogram information. Since a method of selecting a threshold from the histogram information is known, the description thereof is omitted here. The binarization parameter is selected based on the definition body information, depending on whether the use of the information in the item area is for display, for character recognition, or both. Is a parameter for The binarization parameter is, for example, a parameter for adjusting the density of image data at the time of the binarization processing, and this value is defined according to the application. Therefore, when data of a certain item is used for two kinds of uses for display and recognition, two kinds of binarization parameters are prepared for each use.
[0025]
The binarization processing unit 8 performs a binarization process on the image data of the item area using the threshold value and the binarization parameter selected by the threshold value selection unit 7. When a plurality of binarization parameters are set by the threshold value selecting means 7, the binarization processing is performed using the respective binarization parameters.
[0026]
The operation of the image processing apparatus having such a configuration will be described.
The definition body information storage unit 4 stores, in advance, definition body information that defines the position, range, use, and the like of the items constituting the form for the form to be processed.
[0027]
When image data of a predetermined form is input from the image scanner 1 to the image processing apparatus 2, the identification information reading means 3 reads the form identification information described at a predetermined position on the form, and searches for the definition body information. Send to 5. The definition body information searching means 5 searches the definition body information storage unit 4 based on the form identification information, extracts the definition body information corresponding to the target form, stores it in the binarized information storage unit 6, and Deliver to value selection means 7. The threshold value selecting means 7 extracts image data of the item area for each item of the form based on the definition body information, creates histogram information, and selects an optimum threshold value. Further, a binarization parameter is set based on the definition body information. The threshold value and the binarization parameter are stored in the binarization information storage unit 6 and passed to the binarization processing unit 8. The binarization processing unit 8 performs binarization processing of the image data of the item area for each item based on the binarization information stored in the binarization information storage unit 6. When a plurality of binarization parameters are set, the binarization processing is performed using the respective binarization parameters.
[0028]
Next, an embodiment to which the present invention is applied will be described. FIG. 2 is a hardware configuration diagram of the image reading device and the image scanner according to the embodiment of the present invention. The same components as those in FIG. 1 are denoted by the same reference numerals, and description thereof is omitted.
[0029]
The entire image processing apparatus is controlled by a CPU (Central Processing Unit) 12. The memory 13, the image memory 14, the auxiliary storage device 15, and the capture board 11 are connected to the CPU 12 via the bus 16.
[0030]
The capture board 11 is connected to the image scanner 1, captures image data read by the image scanner 1, and stores the image data in the image memory 14. The memory 13 temporarily stores at least a part of a program executed by the CPU 12 and various data necessary for processing by the CPU 12. The image memory 14 stores image data input via the capture board 11. The auxiliary storage device 15 stores an OS (Operating System) and application programs.
[0031]
With the above hardware configuration, the processing functions of the embodiment can be realized.
Hereinafter, the embodiment will be described with reference to a form used in a bank as an example. FIG. 3 is a diagram illustrating a form on which a reading process is performed by the image processing apparatus according to the embodiment of the present invention.
[0032]
The form 100 to be read in the embodiment includes a form ID 110 for identifying the form, a guide character string 120 indicating an item, a recognition field 130 for character recognition, a display field 140 for display data generation, And a seal field 150 on which the seal is stamped.
[0033]
The form ID 110 is a code for identifying a form printed at a predetermined position on the form, and is located at the upper left of the form 100 in the example of FIG. IDs are assigned such as “0101” for the deposit form and “0102” for the withdrawal form.
[0034]
The guide character string 120 is a character that is printed in advance to indicate the content of the item, and in the example of the figure, “bank name”, “branch name”, “recipient”, “address”, “client”, “Address”, “Account number”, and “Amount” are printed.
[0035]
The recognition field 130 is a field in which a character string entered by a customer with handwritten characters with reference to the guide character string 120 is set, and is a target of character recognition. In the example of the figure, a field in which characters corresponding to “bank name”, “branch name”, “payee”, “client”, “account number”, and “amount” are entered is the recognition field 130.
[0036]
The display field 140 is a field for generating image data to be displayed on the display screen for the purpose of confirmation by the operator. In the example of the figure, the guide character string 120 and the entry character of the "address" of the recipient, the "address" of the client, the "account number", and the guide character string 120 and the entry character of the "amount" are displayed in the display field 140. included.
[0037]
The seal field 150 is a field where the seal is pressed.
The processing procedure executed by the image processing apparatus according to the present invention for such a form 100 will be sequentially described.
[0038]
Image data of the form 100 read by the image scanner 1 is input to the image processing apparatus, and a form record acquisition process as definition information is started.
FIG. 4 is a diagram showing a processing procedure for obtaining a form record in the image processing apparatus according to the embodiment of the present invention. The same components as those in FIG. 3 are denoted by the same reference numerals, and description thereof is omitted.
[0039]
When the image data of the form 100 is input, a recognition process of a form ID 110 which is information for identifying the form is performed (step S1). In the recognition process of the form ID 110, the image data of the form ID 110 printed at a predetermined position is extracted, binarized, and then subjected to character recognition to read the form ID.
[0040]
Subsequently, using the form ID obtained in the form ID recognition process (step S1), a process of acquiring a form record as definition body information of the form is performed (step S2). In the form record acquisition process, a form record (definition body information) stored in the form record storage unit 41 is searched using the form ID as a key, and a corresponding form record is extracted. The information of the extracted form record is transferred to the binarized record information file 200.
[0041]
Here, the binarized record information file 200 will be described. FIG. 5 is a configuration diagram of the binarized record information file according to the embodiment of this invention.
The binarized record information file 200 includes, for each item (field name), an information item relating to a range of a field of coordinates 201, a field width 202, and a field height 203 of the item area (hereinafter, referred to as a field) on a form. And information items indicating the use of the fields of the recognition field 204, the display field 205, and the seal field 206 are extracted from the form record and transcribed. Further, the binarized record information file 200 includes information items of a binarization threshold 207 and a binarization parameter 208 for binarization generated in a process described later.
[0042]
In the example of FIG. 3, each of the recognition field 130, the display field 140, and the seal field 150 is rectangular. The range of the field is specified by the coordinates of one point of the rectangle (two-dimensional with X in the horizontal direction and Y in the vertical direction) and the field width in the horizontal direction and the field height in the vertical direction based on these coordinates. These values are described in portions corresponding to each field of the coordinate 201 column, the field width 202 column, and the field height 203 column. In the example of the figure, the fields are classified into any of the recognition field 130 for character recognition, the display field 140 for display, and the seal field 150 for seal. In the recognition field 204 column, display field 205 column, and seal field 206 column provided corresponding to each field, a mark (○) is set when applicable.
[0043]
For example, the field "bank name" exists in a range of 750 dots wide and 75 dots high from the position of the coordinates (X = 200, Y = 180). The usage is a recognition field. In some cases, the field may serve as both a recognition field and a display field. In the example shown in the figure, the fields of “account number” and “money” are recognition / display fields for recognition and display. The column 204 and the display field column 205 are marked.
[0044]
With the above processing procedure, the form record information of the target form is set as the binarized record information, and the image data cutout processing of the field is started based on the binarized record information. FIG. 6 is a diagram illustrating a processing procedure for extracting image data in the image processing apparatus according to the embodiment of the present invention. 3 and 5 are denoted by the same reference numerals, and description thereof is omitted.
[0045]
After the form record information is set in the binarized record information file 200, the image data of the form 100 is obtained by referring to the cutout information of the coordinates 201, the field width 202 and the field height 203 for specifying the field range on the form. The image data in the field range is cut out (step S3). The extracted image data is stored in the work image memory 300.
[0046]
In the example of the figure, cutout information (coordinates, field width, field height) corresponding to the field of “bank name” is extracted from the binarized record information, and the image of the “bank name” field is extracted based on the cutout information. The data, “Fujitsu Ginkgo”, is cut out to the work image memory 300.
[0047]
Next, a threshold value of the field is calculated from the extracted image data. FIG. 7 is a diagram illustrating a processing procedure for calculating a threshold value in the image processing apparatus according to the embodiment of the present invention.
[0048]
In the threshold value calculation processing, first, histogram information representing the distribution of the density levels of the cut-out image data read out to the work image memory 300 is generated (step S4). Next, referring to the information in the recognition field 204, the display field 205, and the seal field 206 representing the field usage of the binarized record information file 200, the usage of the field is determined (step S5). Next, the histogram information created in step S4 is automatically determined, and a threshold value is calculated (step S6). The calculated threshold value is registered in the binarization threshold value 207 column of the corresponding field of the binarization record information file 200. Further, based on the field information determination in step S5, a binarization parameter registered in advance according to the use of the field is selected (step S7). The binarization parameters are set in advance for each application and stored in the memory. In step S7, the corresponding binarized parameter value is read based on the field usage, and registered in the binarized parameter value 208 column of the corresponding field of the binarized record information file 200.
[0049]
In the example of the figure, the threshold value 155 calculated from the image data corresponding to the “bank name” field is set to the binarized threshold value 207 corresponding to the “bank name” field. Further, since there is a mark in the recognition field 204a corresponding to the "bank name" field, it is determined that this field is a recognition field, and the corresponding binarization parameter 0.99 corresponds to the corresponding binarization parameter 0.99. This is set in the column 208a. In the case where the field serves as both a recognition field and a display field, such as the “account number” field in the example of the figure, binarized parameter fields 209a and 209b corresponding to each field use are provided. A binarization parameter is selected and registered for each.
[0050]
By the above procedure, the threshold value and the binarization parameter for each field are calculated based on the form record, and stored in the binarization record information file.
Next, field binarization processing is performed based on the set binarized record information file. FIG. 8 is a diagram showing a binarization processing procedure in the image processing apparatus according to the embodiment of the present invention.
[0051]
In the binarization processing, the image data of the form is sequentially scanned using the coordinates, and the binarization processing is performed for each field.
The current coordinate being scanned is compared with the value in the coordinate 201 column of the binarized record information file 200 to detect a field (step S9). If a field is detected, the field usage is determined with reference to the recognition field 204 column, display field 205 column, and seal field 206 column of the binarized record information file 200 (step S10). Next, the binarization threshold and the binarization parameter of the corresponding field are read from the binarization threshold 207 column and the binarization parameter 208 column of the binarization record information file 200, and the binarization process is performed. A threshold and a binarization parameter are acquired (step S11). Then, a binarization process is performed using the obtained threshold value and the binarization parameter (step S12). As described above, when the field is detected, the threshold value and the binarization parameter are switched to the values set in accordance with the field, and the binarization process is performed.
[0052]
As described above, the image processing apparatus of the present invention calculates the threshold value of the limited area in the field based on the form record information, and performs the binarization processing of this field. The processing is performed by switching to the calculated threshold value. Further, the binarization parameter is switched according to the use of the field.
[0053]
As described above, since the calculation is performed in a limited area in the field, the threshold value can be obtained without being affected by the ratio of the form color / guide character / black printing. Since the binarization of the field is performed by switching to the threshold value calculated in this way, loss of information in the field can be minimized. Further, by recognizing the use of the field and selecting a binarization parameter, it is possible to perform a binarization process according to each use. As a result, clearer display is possible in the case of a display field, and the recognition rate of characters is improved in the case of a recognition field.
[0054]
Next, image processing executed by a computer using the image processing program according to the present invention will be described. By causing the computer to execute the image processing program, the computer functions as the image processing device according to the present invention.
[0055]
FIG. 9 is a flowchart showing image processing according to the embodiment of the present invention.
The image processing according to the present invention starts after a form is read by the image scanner 1.
[Step S21] Image data read by the image scanner is input and signal-processed, and a process of reading a form ID from the signal-processed image data is performed.
[Step S22] A process of acquiring definition body information based on the form ID is performed.
[Step S23] Based on the acquired definition body information, a process of calculating a threshold value (and a binarization parameter) for each field is performed.
[Step S24] The form is binarized by switching to the threshold value (and binarization parameter) calculated for each field.
[0056]
Each process will be described in order.
The image data input and form ID reading processing (step S21) will be described. FIG. 10 is a flowchart showing a processing procedure of image data input and form ID reading processing according to the embodiment of the present invention. The image scanner 1 inputs image data obtained by capturing the form, and the processing is started.
[Step S211] Optical image data acquired by the CCD element of the image scanner is read by the image scanner.
[Step S212] Shading correction (shadow correction) is performed on image data read by the image scanner to remove shadows.
[Step S213] The image data signal is divided into R, G, and B, and a luminance component Y signal is generated.
[Step S214] Using the luminance component Y signal, the image data of the form ID of the form is cut out and binarized.
[Step S215] Image data of the binarized form ID is subjected to character recognition (OCR recognition), and a form ID is acquired.
[0057]
Through the above processing procedure, the image data and the form ID are obtained.
Next, the definition body information acquisition processing (step S22) will be described. FIG. 11 is a flowchart illustrating a processing procedure of the definition body information acquisition processing according to the embodiment of this invention. After the form ID is obtained, the process is started.
[Step S221] A form ID recognition result is acquired, and the form record information storage unit is searched based on the acquired form ID, and a form record (definition body) of the corresponding form is extracted.
[Step S222] Binarized record information for binarization is created.
[Step S223] The form record extracted in step S221 is copied to a corresponding column of the binarized record information. By the copy, the coordinates indicating the range of the field, the width and height of the field, and information indicating whether the use of the field corresponds to the recognition field, the display field, or the seal field are set in the binary record information file.
[0058]
According to the above processing procedure, information on the range of the field and information on the use of the field are set in the binarized record information file.
Next, the threshold value calculation processing (step S23) will be described. FIG. 12 is a flowchart illustrating a processing procedure of the threshold value calculation processing according to the embodiment of this invention. After the field range information (coordinates, width, height) and application information (recognition, display, and seal) are set in the binarized record information file, the process starts.
[Step S231] With reference to field range information (coordinates, field width and height) of the binarized record information file, image data (image) of the field is cut out.
[Step S232] Histogram information of the extracted image data (image) is created, and a threshold value is obtained based on the histogram information.
[Step S233] The calculated threshold is reflected in the binarized record information. The calculated threshold value is registered in the binarization threshold value column corresponding to the target field.
[Step S234] A binarization parameter is selected according to the field usage of the binarized record information file, and set in a corresponding column of the binarized record information file. When there are a plurality of field uses, a binarization parameter corresponding to each use is selected and set.
[Step S235] It is determined whether or not there is information on the next field in the binarized record information file. If there is no field, the process ends. If the next field exists, the process returns to step S231 to repeat the process for the next field.
[0059]
By the above processing, the binarization threshold value and the binarization parameter corresponding to each field are calculated and set in the binarization record information file.
Next, the binarization process (Step S24) will be described. FIG. 13 is a flowchart showing a processing procedure of the binarization processing according to the embodiment of the present invention. After the thresholds and the binarization parameters of all the fields are set in the binarization record information file, the process starts.
[Step S241] An image pointer indicating coordinates for performing the binarization process is set at the upper left of the form. The binarization process scans a form according to an image pointer and is sequentially executed.
[Step S242] The field range information of the binarized record information file is compared with the image pointer, and it is determined whether the location pointed to by the image pointer is a recognition / display field. A combination of the recognition field and the display field is referred to as a recognition / display field. If not, the process advances to step S244.
[Step S243] If the instruction of the image pointer is a recognition / display field, a binarized record information file corresponding to this field is specified, and the recognition / display field is specified and the designated field binarization processing is executed. After performing the binarization process, the process proceeds to step S251.
[Step S244] If it is not a recognition / display field, it is determined whether the location pointed to by the image pointer is a recognition field. If it is in the recognition field, the process proceeds to step S246.
[Step S245] If the instruction of the image pointer is a recognition field, a binarized record information file corresponding to this field is specified, a recognition field is specified, and a designated field binarization process is executed. After performing the binarization process, the process proceeds to step S251.
[Step S246] If it is not a recognition field, it is determined whether the location pointed to by the image pointer is a display field. If it is not a display field, the process proceeds to step S248.
[Step S247] If the indication of the image pointer is a display field, a binarized record information file corresponding to this field is indicated, a display field is designated, and a designated field binarization process is executed. After performing the binarization process, the process proceeds to step S251.
[Step S248] If it is not a display field, it is determined whether the location pointed to by the image pointer is a seal field. If not, the process advances to step S250.
[Step S249] If the instruction of the image pointer is a seal field, a binarized record information file corresponding to this field is specified, and the seal field is specified to execute the designated field binarization processing. After performing the binarization process, the process proceeds to step S251.
[Step S250] If the field is not the field specified in the binarized record information, the binarizing process is performed using the threshold value of the entire form.
[Step S251] The image pointer is moved. When the processing of the field is completed, the image pointer moves to the last end of the field. For example, when scanning is sequentially performed in the horizontal direction from the upper left, the image pointer moves from the upper left to the upper right of the field.
[Step S252] It is determined whether the location pointed to by the image pointer is the end of the image data (image). If it is the last point of the image data, the process ends. If it is not the last point, the process returns to step S242 to perform the process.
[0060]
Here, the binarization processing of the designated field will be described.
FIG. 14 is a flowchart illustrating a processing procedure of the designated field binarization processing according to the embodiment of this invention. The binarization information file corresponding to the field is specified, and the process starts.
[Step S261] A threshold value is obtained from the corresponding binarized record information.
[Step S262] A binarization parameter is obtained from the corresponding binarization record information.
[Step S263] The acquired threshold value and binarization parameter are set.
[Step S264] Binarization processing of the designated field is performed using the set threshold value and the binarization parameter.
[Step S265] It is determined whether the designated field is a recognition / display field. If it is not a recognition / display field, the process ends.
[Step S266] If the designated field is a recognition / display field, since there are two types of binarization parameters, it is determined whether or not processing using all binarization parameters has been completed. If all the parameters have been completed, the process ends. If all the parameters have not been completed, the process returns to step S263, and the procedure of performing the binarization process using another binarization parameter is repeated.
[0061]
As described above, according to the image processing method of the present invention, the optimum threshold value and the binarization parameter are set for each field, and the binarization processing is performed for each field based on the threshold value and the binarization parameter. Is performed, the loss of information can be minimized.
[0062]
Note that the above processing functions can be realized by a computer. In that case, a program is provided that describes the processing contents of the functions that the image processing apparatus should have. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing content can be recorded on a computer-readable recording medium. Computer-readable recording media include magnetic recording devices, optical disks, magneto-optical recording media, and semiconductor memories. The magnetic recording device includes a hard disk device (HDD), a flexible disk (FD), a magnetic tape, and the like. Examples of the optical disc include a DVD (Digital Versatile Disc), a DVD-RAM (Random Access Memory), a CD-ROM (Compact Disc Read Only Memory), and a CD-R (Recordable) / RW (ReWritable). The magneto-optical recording medium includes an MO (Magneto-Optical disk) and the like.
[0063]
When distributing the program, for example, portable recording media such as DVDs and CD-ROMs on which the program is recorded are sold. Alternatively, the program may be stored in a storage device of a server computer, and the program may be transferred from the server computer to another computer via a network.
[0064]
The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, the computer may execute the processing according to the received program each time the program is transferred from the server computer.
[0065]
(Supplementary Note 1) In an image processing apparatus that performs image processing for binarizing image information of a form optically read using a predetermined threshold value,
Identification information reading means for reading form identification information for identifying the form from the image information,
Definition body information storage means for storing definition body information including a position and a range of the item in the form defined for each item constituting the form,
A definition body information search unit that searches the definition body information storage unit based on the form identification information and extracts the definition body information corresponding to the form;
A threshold value selecting unit that selects a threshold value according to the density level of the image information of the item area defined for each item based on the definition body information;
Binarizing means for performing binarization processing of image information of the item area using the threshold value for each of the items;
An image processing apparatus comprising:
[0066]
(Supplementary Note 2) The threshold value selecting means further includes:
2. The image processing apparatus according to claim 1, wherein a use of the item is extracted from the definition body information, and a binarization parameter for adjusting the binarization process defined according to the use is selected.
[0067]
(Supplementary Note 3) The item area may be set according to the use.
A display field for using the image information of the item area for display,
A recognition field for performing character recognition of the image information of the item area;
3. The image processing apparatus according to claim 2, wherein the image processing apparatus is classified into:
[0068]
(Supplementary Note 4) The image processing apparatus according to Supplementary Note 2, wherein the threshold value selecting unit selects the binarization parameter for each of the uses when the item is used for a plurality of uses.
[0069]
(Supplementary Note 5) When the plurality of binarization parameters are selected in accordance with the application for the item, the binarization unit includes, for each of the binarization parameters, the binarization parameter and the threshold value. 3. The image processing apparatus according to claim 2, wherein a binarization process of the image information of the item area is performed using the image data.
[0070]
(Supplementary Note 6) The threshold value selecting unit creates histogram information of the image information of the item area defined by the definition body information, and selects the threshold value based on the histogram information. 2. The image processing apparatus according to claim 1, wherein
[0071]
(Supplementary Note 7) In an image processing program for image processing for binarizing image information of a form read optically with a predetermined threshold value,
On the computer,
Storing in a predetermined storage means definition body information including a position and a range of the item in the form defined for each item constituting the form;
Reading form identification information for identifying the form from the image information,
Based on the form identification information, searching the predetermined storage means in which the definition body information is stored, extracting the definition body information corresponding to the form,
Selecting a threshold value according to the density level of the image information of the item area defined for each item based on the definition body information;
Performing binarization processing of image information of the item area using the threshold value for each of the items;
An image processing program characterized by executing
[0072]
(Supplementary Note 8) In an image processing method for binarizing image information of a form read optically with a predetermined threshold value,
Storing in a predetermined storage means definition body information including a position and a range of the item in the form defined for each item constituting the form;
Reading form identification information for identifying the form from the image information,
Based on the form identification information, searching the predetermined storage means in which the definition body information is stored, extracting the definition body information corresponding to the form,
Selecting a threshold value according to the density level of the image information of the item area defined for each item based on the definition body information;
Performing binarization processing of image information of the item area using the threshold value for each of the items;
An image processing method comprising:
[0073]
【The invention's effect】
As described above, in the present invention, a form is identified, and based on the identified information, definition body information that is defined for the items constituting the form is obtained. Then, a threshold value optimized for each item is selected based on the definition body information, and binarization processing is performed on the image information of the corresponding item region using the threshold value.
[0074]
As described above, since the binarization process is performed using the threshold value optimized for each item, loss of information can be minimized. As a result, if the item is a character recognition area, the accuracy of character recognition can be improved.
[0075]
Further, by selecting a binarization parameter defined in accordance with the use of the item based on the definition body information and performing binarization processing using the binarization parameter, the loss of information is further reduced. Becomes possible.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram of the invention applied to an embodiment of the present invention.
FIG. 2 is a hardware configuration diagram of an image reading apparatus and an image scanner according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating a form on which a reading process is performed by the image processing apparatus according to the embodiment of the present invention;
FIG. 4 is a diagram illustrating a processing procedure for acquiring a form record in the image processing apparatus according to the embodiment of the present invention.
FIG. 5 is a configuration diagram of a binarized record information file according to the embodiment of this invention.
FIG. 6 is a diagram illustrating a processing procedure for extracting image data in the image processing apparatus according to the embodiment of the present invention.
FIG. 7 is a diagram illustrating a processing procedure for calculating a threshold value in the image processing apparatus according to the embodiment of the present invention.
FIG. 8 is a diagram illustrating a binarization processing procedure in the image processing apparatus according to the embodiment of the present invention.
FIG. 9 is a flowchart illustrating image processing according to the embodiment of the present invention.
FIG. 10 is a flowchart showing a processing procedure of image data input and form ID reading processing according to the embodiment of the present invention.
FIG. 11 is a flowchart illustrating a processing procedure of a definition body information acquisition processing according to the embodiment of this invention.
FIG. 12 is a flowchart showing a processing procedure of a threshold value calculation processing according to the embodiment of the present invention.
FIG. 13 is a flowchart showing a processing procedure of a binarization process according to the embodiment of the present invention.
FIG. 14 is a flowchart illustrating a processing procedure of a designated field binarization processing according to the embodiment of this invention.
FIG. 15 is a diagram illustrating an example of a form.
[Explanation of symbols]
1 Image scanner
2 Image processing device
3 identification information reading means
4 Definition body information storage
5 Definition body information search means
6 Binary information storage unit
7 Threshold value selection means
8 Binarization processing means

Claims (5)

An image processing apparatus for performing image processing for binarizing image information of a form optically read using a predetermined threshold value,
Identification information reading means for reading form identification information for identifying the form from the image information,
Definition body information storage means for storing definition body information including a position and a range of the item in the form defined for each item constituting the form,
A definition body information search unit that searches the definition body information storage unit based on the form identification information and extracts the definition body information corresponding to the form;
A threshold value selecting unit that selects a threshold value according to the density level of the image information of the item area defined for each item based on the definition body information;
Binarizing means for performing binarization processing of image information of the item area using the threshold value for each of the items;
An image processing apparatus comprising: The threshold value selecting means further comprises:
2. The image processing apparatus according to claim 1, wherein a use of the item is extracted from the definition body information, and a binarization parameter for adjusting the binarization process defined according to the use is selected. The image processing apparatus according to claim 2, wherein when the item is used for a plurality of uses, the threshold value selecting unit selects the binarization parameter for each use. The binarization unit, when a plurality of the binarization parameters are selected for the item according to the application, uses the binarization parameter and the threshold value for each of the binarization parameters. 3. The image processing apparatus according to claim 2, wherein the image information of the item area is binarized. An image processing program for image processing for binarizing image information of a form optically read at a predetermined threshold value,
On the computer,
Storing in a predetermined storage means definition body information including a position and a range of the item in the form defined for each item constituting the form;
Reading form identification information for identifying the form from the image information,
Based on the form identification information, searching the predetermined storage means in which the definition body information is stored, extracting the definition body information corresponding to the form,
Selecting a threshold value according to the density level of the image information of the item area defined for each item based on the definition body information;
Performing binarization processing of image information of the item area using the threshold value for each of the items;
An image processing program characterized by executing

Images