JP2000293596A - Slip recognition device, information processing system, slip recognition method and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely reflect the similarity of cell distributions in the comparison processing of the table format detailed structures that is carried out for recognizing a slip and accordingly to improve the slip recognizing accuracy by using directly the data which are outputted from an image feature value extraction means as the comparison data and performing the comparison of cells through the comparison of number of pieces of line units. SOLUTION: In this slip recognizer, a processor 11 has an image feature value extraction means 11a which extracts the feature value of the image data on slips that is read by a scanner 13, a character recognition means 11b which recognizes the character areas included in the image data and similarity calculation means 11c which calculates the similarity of slips. Then the processor 11 carries out the arithmetic processing in response to various types of programs and also carries out various types of processing necessary for recognizing the slips. In this device, the comparison data mean the cross coordinate value of the table ruled lines and use directly the data which are outputted from the means 11a. As a result, no error is caused for production of the format information. Furthermore, the variance of similarity due to the cell distribution situation can be reduced since the comparison of numbers of pieces of cells is carried out through the comparison of the number of pieces of line units.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a form recognition apparatus and form recognition method for automatically recognizing forms, an information processing system including the form recognition apparatus, and a storage medium for implementing the form recognition method. .

[0002]

2. Description of the Related Art A form recognizing apparatus capable of automatically classifying each form when processing a large number of forms has been known.

In this type of form recognition apparatus, the position of a cell in a form table and the number of cells have been used as comparison data in comparison processing of the detailed structure of the table format in the form. That is, a method using the center coordinates of a cell or a method called a grid division method is used to determine the degree of coincidence of cell positions.

A method of using the center coordinates of a cell is to calculate the center coordinates from the square coordinate values of the cell obtained by the image extraordinary amount extraction means and compare the center coordinates to determine the degree of coincidence of the cell position. judge. On the other hand, in the grid division method, the entire table is divided into a grid by a plurality of grids, and the number of intersections of table rule lines included in each block created by this is compared.

As a method of comparing the number of cells, a method of simply comparing the total number of cells in a table is generally known.

[0006]

However, the comparison processing of the detailed table format structure in the conventional form recognition apparatus has the following problems.

(1) In the above-described method using the center coordinates of the cell, it is necessary to calculate the center coordinates from the coordinate values of the cell rectangle obtained by the extra-trace extraction means of the image, and an error occurs in the calculation process. I do. That is, the table format data itself contains an error.

(2) The grid division method is a method for obtaining a table format in the image feature extraction means. The grid division method requires more processing than the method using coordinate values of feature points obtained by the image feature extraction means. It's complicated.

(3) In comparison of the number of cells in the entire table, if the total number of cells is the same even if the cell distribution is different, it is determined that the tables are similar, and thus there is a problem in the determination accuracy.

The present invention has been made in view of the above-described conventional problems, and has been described in comparison processing of a detailed table format structure in form recognition.
It is an object of the present invention to provide a form recognition device, an information processing system, a form recognition method, and a storage medium that improve form recognition accuracy by more closely reflecting the similarity of cell distribution in calculation of similarity.

[0011]

According to a first aspect of the present invention, there is provided a form recognition apparatus for extracting a very small amount of form image data.
A form format data creating unit for generating form format data based on line information obtained by the image feature amount extracting unit, and a form recognition unit for performing form recognition based on the form format data.

According to a second aspect of the present invention, there is provided the form recognition apparatus according to the first aspect, further comprising an image input unit for reading the form as image data, wherein the image feature amount extracting unit includes the image input unit. The feature amount of the form image data read by the means is extracted.

According to a third aspect of the present invention, there is provided the form recognition apparatus according to the first or second aspect, further comprising a storage unit for storing the form format data.

According to a fourth aspect of the present invention, in the form recognition apparatus according to any one of the first to third aspects, the form format data creating means includes a form image extracted by the image extra-volume extracting means. Means for generating registration table format data based on intersection coordinate values of table ruled lines and cells in the data, wherein the storage means stores the registration table format data together with a code for identifying the registration table format data. I do.

In the form recognition apparatus according to a fifth aspect of the present invention, in the form recognition apparatus according to the fourth aspect, the form format data creating means includes a table ruled line in the form image data extracted by the image extra amount extracting means. Means for generating search table format data based on the intersection coordinate values of the cells and the cells, wherein the form recognition means compares the search table format data with the registration table format data stored in the storage means, and performs the search. A similarity calculating means for calculating a similarity between the table format data and the registration table format data is provided.

In the form recognition apparatus according to a sixth aspect of the present invention, in the form recognition apparatus according to the fifth aspect, a similarity to the search table format data is obtained based on a calculation result obtained by the similarity calculation means. Form recognition result output means for outputting the identification code of the registration table format data having a high degree and the similarity as the form recognition result is provided.

According to a seventh aspect of the present invention, in the information processing system having a form recognition apparatus for automatically recognizing a form, the form recognition apparatus includes an image feature amount for extracting a very small amount of form image data. Extracting means, form form data creating means for generating form format data based on line information obtained by the image feature quantity extracting means, and form recognizing means for performing form recognition based on the form format data. And

According to an eighth aspect of the present invention, in the information processing system according to the seventh aspect, the form recognition device includes image input means for reading the form as image data, and the image feature amount extraction means. The means is configured to extract a feature amount of the form image data read by the image input means.

According to a ninth aspect of the present invention, in the information processing system according to the seventh or eighth aspect, the form recognition device has a storage unit for storing the form format data. And

According to a tenth aspect of the present invention, in the information processing system according to any one of the seventh to ninth aspects, the form format data creating means includes a form image extracted by the image extra-quantity extracting means. Means for generating registration table format data based on intersection coordinate values of table ruled lines and cells in the data, wherein the storage means stores the registration table format data together with a code for identifying the registration table format data. I do.

In the information processing system according to an eleventh aspect of the present invention, in the information processing system according to the tenth aspect, the form format data creating means includes a table ruled line in the form image data extracted by the image extra amount extracting means. Means for generating search table format data based on the intersection coordinate values of the cells and the cells, wherein the form recognition means compares the search table format data with the registration table format data stored in the storage means, and performs the search. A similarity calculating means for calculating a similarity between the table format data and the registration table format data is provided.

In the information processing system according to a twelfth aspect of the present invention, in the information processing system according to the eleventh aspect, similarity to the search table format data is obtained based on a calculation result obtained by the similarity calculating means. Form recognition result output means for outputting the identification code of the registration table format data having a high degree and the similarity as the form recognition result is provided.

According to a thirteenth aspect of the present invention, there is provided a form recognition method for extracting an image feature amount for extracting a very small amount of form image data, and generating form format data based on line information obtained by the image feature amount extraction process. Form form data creation processing and form recognition processing for performing form recognition based on the form format data are executed.

According to a fourteenth aspect of the present invention, in the form recognition method according to the thirteenth aspect, an image input process for reading the form as image data is performed, and the image feature amount extracting process is performed in the image input process. The feature amount of the form image data read by the above is extracted.

According to a fifteenth aspect of the present invention, there is provided the form recognition method according to the thirteenth or fourteenth aspect, further comprising a storage process for storing the form format data.

According to a sixteenth aspect of the present invention, in the form recognition method according to any one of the thirteenth to fifteenth aspects, the form format data creation processing is performed on the form image extracted by the image extraordinary amount extraction processing. A process of generating registration table format data based on intersection coordinate values of table ruled lines and cells in the data, wherein the storage process stores the registration table format data together with a code for identifying the registration table format data. I do.

According to a seventeenth aspect of the present invention, in the form recognition method according to the sixteenth aspect, the form format data creation processing is performed using table ruled lines in the form image data extracted by the image extraordinary amount extraction processing. And processing for generating search table format data based on the intersection coordinate values of cells and the cells. The form recognition process compares the search table format data with the registration table format data stored in the storage process, and performs the search. It has a similarity calculation process for calculating the similarity between the table format data and the registered table format data.

In the form recognition method according to the present invention, based on the calculation result obtained in the similarity calculation processing, similarity to the search table format data is obtained. It has a form recognition result output process for outputting the identification code of the registration table format data having a high degree and the similarity as the form recognition result.

A storage medium according to a nineteenth aspect of the present invention is a storage medium storing a computer-readable program for executing a form recognition method for automatically recognizing a form. An image feature amount extraction step of extracting a very small amount of image data; a form format data creation step of generating form format data based on line information obtained by the image feature amount extraction step; and a form recognition based on the form format data. And a form recognition step to be performed.

According to a twentieth aspect of the present invention, there is provided the storage medium according to the nineteenth aspect, further comprising an image input step of reading a form as image data, wherein the image feature amount extracting step is performed by the image input step. The feature amount of the read form image data is extracted.

A storage medium according to a twenty-first aspect of the present invention is the storage medium according to the nineteenth or twentieth aspect, further comprising a storage step of storing the form format data.

In the storage medium according to a twenty-second aspect of the present invention, in the storage medium according to any one of the nineteenth to twenty-first aspects, the form format data creating step includes a step of creating the form image data extracted in the image extra-trace extraction step. The method further comprises the step of generating registered table format data based on the intersection coordinate values of table ruled lines and cells, and wherein the storing step stores the registered table format data together with a code for identifying the registered table format data.

[0033] In the storage medium according to the twenty-third aspect of the present invention, in the storage medium according to the twenty-second aspect, the form format data creating step includes a table ruled line and a cell in the form image data extracted in the image extra-trace extraction step. Generating the search table format data based on the intersection coordinate values of the search table format, wherein the form recognition step compares the search table format data with the registration table format data stored in the storage step to generate the search table format data. A similarity calculating step of calculating a similarity between the data and the registration table format data.

According to a twenty-fourth aspect of the present invention, in the storage medium according to the twenty-third aspect, based on a calculation result obtained in the similarity calculation step, a similarity is calculated with respect to the search table format data. A form recognition result output step of outputting the identification code of the high registration table format data and the similarity as a form recognition result is provided.

[0035]

Embodiments of the present invention will be described below in detail with reference to the drawings.

[First Embodiment] FIG. 1 is a block diagram showing a schematic configuration of a form recognition apparatus according to a first embodiment of the present invention.

This form recognition apparatus includes a processor 11 for controlling the operation of the entire apparatus, a memory 12 for storing various data, a scanner 13 for reading a form as image data, and an instruction operation for registering and retrieving a form. A keyboard 14 for performing the operation, a disk 15 functioning as an auxiliary storage device, and a display 16 for displaying various information such as a recognition result are connected to each other via a bus 20.

The processor 11 includes an image feature extracting means 11a for extracting a feature of image data of a form read by the scanner 13, a character recognizing means 11b for recognizing a character area in the image data, and a flowchart shown in FIG. Similarity calculating means 11c for calculating the similarity of a form according to
And performs arithmetic processing according to various control programs,
Executes various processes required for form recognition.

The image read by the scanner 13 is binarized and sent to the image feature extraction unit 11a. The image is divided into regions such as tables, texts, and pictures by a method such as a black dot histogram method. For example, in the case of a table block, a detailed structure of the table is obtained by a ruled line tracking method or the like. If it is a text block, it is converted to a character code by character recognition means.

In the information thus obtained, the detailed structure of the cells of the table is stored in the memory 12 and the disk 15 as table format data 12a shown in FIGS. 2 (a) and 2 (b).

In the example of the line format shown in FIG. 2A, first, a cell whose upper left corner is in contact with the horizontal imaginary ruled line at the table origin is defined as a cell of the 0th line. Next, the horizontal virtual ruled line in the direction below the Y axis is defined as the first line, and similarly, the next horizontal virtual ruled line is defined as the second line. Each line information is shown in FIG.
As shown in (b), the number of line information, the starting X coordinate of the line (the left end position of the horizontal ruled line), and the lower right corner X coordinate of the cell of the line are recorded. Apart from this information, a table
The virtual ruled line position from the top position is recorded as line position information.

Using these table format data,
Recognize tables in forms. The form recognition system receives commands such as form registration and form search from the keyboard 14, and the processor 11 performs corresponding processing by using the table format data 12a. And
The result is displayed on the display 16.

The operation of the form recognition system according to the present embodiment, in particular, various control processes executed by the processor 11 will be described below with reference to FIGS.

FIGS. 3A and 3B are diagrams showing sample tables for obtaining similarity in the processing procedure of the present embodiment. FIG. 3A shows a search table, and FIG. Shows a table.

The degree to which the search table is similar to the registration table is determined, and the most similar registration table is regarded as a table recognized by the search table.

FIG. 4 is a flowchart showing a procedure for calculating the similarity according to the present embodiment. The similarity calculation program according to this flowchart is stored in the memory 12
Alternatively, it is stored on the disk 15.

First, in step S11, the average value of the width and height of the search table and the registration table is obtained. Search table width = 4-1, search table height = FA, registration table width = 4-1, registration table height = EA
AVX = (width of search table + width of registration table) / 2 AVY = (height of search table + height of registration table)
/ 2. Next, the number of cells of a table having a large number of cells among these tables is set in MAXC. MAXC
= 15.

The processing after step S12 is processing for recording the difference between the two tables as a penalty value.
These penalty values are obtained for each line and added up to the last line.

In step S12, the number of cells recorded in the line information of the line (the first cell in contact with the horizontal virtual ruled line of A) is compared using the following equation.

PC = KC * | (cell position of search table) − (number of cells of search table) | / MAXC The number of line information in the search table is 4, and the number of line information in the registration table is 4. The penalty PC = 0 due to the number of cells. KC is a coefficient for penalty weighting.

In step S13, the position of the cell recorded in the line information of the line is compared. The line information of the search table is {1, 2, 3, 4}, and the line information of the registration table is {1, 2, 3, 4}. | (Search line 1)-(registration line 1) |, | (search line 2)-
(Registration line 2) |, | (search line 3)-(registration line 3) | (search line 4)-(registration line 4) |

The penalty value is calculated by substituting these difference values into the following equation.

PX = KX * | (cell position of search table)-(cell position of registration table closest to cell position of search table) / AVX KX is a coefficient used for penalty weighting.
The difference value is calculated by extracting the line closest to the search line from the registered line information. For example, if the line closest to the search line 2 is the registered line 3, | (search line 2) − (registered line 3) | In this case, the registration line 2 is not used for penalty calculation, but will be located between 1 and 2 of the search line, so that the number of cells in the registration line is likely to increase by one. , And the number of cells is considered to have been added.

In step S14, a penalty value for the difference between the line positions is calculated by the following equation.

PY = KY * | (the relevant line position in the search table)-(the relevant line position in the registration table) | / AV
Y is the line position A of the search table and the line position A of the registration table. Since the origin is at the upper left corner of the table, the line position is 0 in both tables, and the penalty value PY = 0. Note that KY is a weighting coefficient for penalty of the line position.

In step S15, the next line to be compared is calculated. Since the line in the search and registration table is closest to the B line, the line information of the B line is acquired.

In step S17, since the penalty calculation has not yet been performed up to the last line of the table, the process returns to step S12 to calculate the penalty value of the next line and add it to the penalty value of line A.

In step S15, a line to be compared next is obtained. This time, since the D line of the search table and the C line of the registration table are the closest positions, the line information is acquired.

In step S16, since the C line of the search table is skipped, the number 3 of cells in this line is added to the penalty. In step S17, since it is not the last line yet, the process returns to step S12 to restart the process of calculating the penalty value. In this step S17, if the process reaches the last line in either the search or registration table, the process exits the above calculation loop and proceeds to step S17.
Proceeding to 18, the number of cells in the remaining lines is calculated and added to the penalty value.

Then, the final similarity is calculated in step S19. Assuming that the similarity of exactly the same form is 1, subtracting a value obtained by adding all penalty values from 1 makes it possible to acquire the similarity of the search table with respect to the registration table.

As described above, in the form recognition method of this embodiment, line information as shown in FIGS. 2A and 2B is used as a table format. First, a cell in which the upper left corner of the cell touches a horizontal ruled line intersecting the table origin (upper left corner of the table) is defined as a cell on the 0th line. Next, the horizontal ruled line which moves downward and appears first is defined as the first line, and similarly, the second horizontal line is defined downward. Each line stores the number of pieces of information, the X coordinate of the head of the line (the lower left corner position of the head cell), and the X coordinate of the lower right corner of the cell as a line format. In short, in the comparison of the table formats, the line information of the approximate line position is used as the format data. That is, the comparison data is the intersection coordinate value of the table ruled line, and the data output by the image feature means is used as it is, so that no error occurs for creating the format information. Further, since the comparison of the number of cells is performed by comparing the number of lines, the variation of the similarity due to the distribution state of the cells is reduced.

[Second Embodiment] In the above embodiment, a penalty value generated by comparing all table formats is used for calculating similarity. However, considering how to use the form recognition system, out of a large number of registered forms,
A form similar to the search form must be picked up. Considering the recognition speed, it is preferable to calculate the similarity by the following procedure.

That is, a threshold value of the similarity for determining that the form is the same as the search form is determined in advance, and only when a form having the similarity or higher is found, the identification number and the similarity of the form are output. I do.

In order to realize this system, the table format of the search form and the threshold value of the similarity are incorporated as parameters in the procedure of the present invention. That is, if the similarity falls below the threshold value in the procedure of the present invention, it is sufficient to incorporate a procedure for stopping the calculation and executing the calculation of the similarity in the next table format.

As an example, the following procedure is added. In the procedure of step S16 in FIG. 4, the penalty value of the line information of one line has been calculated. In the next step, similarity calculation as shown in step S19 is performed, and if the result is equal to or smaller than the threshold, the process of calculating the similarity of the table may be exited.

The present invention is not limited to the apparatus of the above-described embodiment, and may be applied to a system including a plurality of devices or an apparatus including a single device. A storage medium storing program codes of software for realizing the functions of the above-described embodiments is supplied to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus reads out and executes the program code stored in the storage medium. It goes without saying that it will be completed by doing so.

In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention. As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk, a hard disk, an optical disk,
Magneto-optical disks, CD-ROMs, CD-Rs, magnetic tapes, nonvolatile memory cards, and ROMs can be used. When the computer executes the readout program code, not only the functions of the above-described embodiments are realized, but also the OS or the like running on the computer performs the actual processing based on the instruction of the program code. It goes without saying that a case where some or all of the operations are performed and the functions of the above-described embodiments are realized by the processing is also included.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the program code is read based on the next program code. Needless to say, the extended function may be performed by a CPU or the like provided in an expansion board or an expansion unit to perform a part or all of the actual processing, and the processing may realize the functions of the above-described embodiments. No.

[0069]

As described in detail above, the form recognition apparatus according to the first to sixth aspects of the present invention, the information processing system according to the seventh to twelfth aspects, and the thirteenth aspect.
According to the form recognition method of the present invention and the storage medium of the present invention, in the comparison processing of the table detailed structure, the image extraordinary amount extracting means is output as comparison data. Since the data to be used is used as it is, no error occurs for creating the form data. Further, since the comparison of the number of cells is performed by comparing the number of lines, the variation in similarity due to the distribution of cells is reduced, and the form recognition accuracy can be improved. That is, the line information is used in the comparison processing of the table detailed structure, and the number of cells in the form table can be checked in line units. This makes it possible to reflect the similarity of the cell distribution more finely in the calculation of the similarity than in the simple check of the number of cells in the form table, and to use the registration table format most similar to the search table format. It can be accurately identified.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of a form recognition device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating table format data according to the first embodiment.

FIG. 3 is a diagram showing a sample table for obtaining a similarity in the processing procedure of the first embodiment.

FIG. 4 is a flowchart illustrating a procedure for calculating a similarity according to the embodiment;

[Explanation of symbols]

 Reference Signs List 11 processor 11a image feature extraction means 11b character recognition means 11c similarity calculation means 12 memory 13 scanner 14 keyboard 15 disk 16 display 20 bus

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 15/70 330G 460E F-term (Reference) 5B029 BB02 CC18 CC27 5B075 MM11 ND03 ND06 ND23 NK07 NK13 NK39 NK54 PP02 PP04 PP12 PP30 PQ02 PQ15 PR06 QM08 UU40 5L096 BA20 FA03 FA16 FA53 FA64 FA69 FA73 JA03 KA15

Claims (24)

[Claims] 1. An image feature amount extracting unit for extracting a very small amount of form image data, a form format data generating unit for generating form format data based on line information obtained by the image feature amount extracting unit, and the form format data And a form recognition unit for performing form recognition based on the form. 2. An image input means for reading a form as image data, wherein the image feature quantity extracting means is configured to extract a feature quantity of the form image data read by the image input means. The form recognition device according to claim 1. 3. The form recognition apparatus according to claim 1, further comprising a storage unit for storing the form format data. 4. The form format data creating means has means for generating registration table format data based on intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature extraction means. 4. The form recognition apparatus according to claim 1, wherein the means stores the registration table format data together with a code for identifying the registration table format data. 5. The form format data generating means includes means for generating search table format data based on intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature amount extracting means. Recognition means,
A similarity calculating unit that compares the search table format data with the registration table format data stored in the storage unit and calculates a similarity between the search table format data and the registration table format data. The form recognition device according to claim 4, wherein 6. An identification code of registration table format data having a high similarity to the search table format data and the similarity are output as a form recognition result based on the calculation result obtained by the similarity calculation means. 6. The form recognition apparatus according to claim 5, further comprising a form recognition result output unit. 7. An information processing system having a form recognizing device for automatically recognizing a form, wherein the form recognizing device comprises: an image feature amount extracting unit for extracting a very small amount of form image data; and the image feature amount extracting unit. An information processing system comprising: a form format data generating unit configured to generate form format data based on obtained line information; and a form recognition unit configured to perform form recognition based on the form format data. 8. The form recognizing device includes image input means for reading a form as image data, and the image feature quantity extracting means extracts a feature quantity of the form image data read by the image input means. The information processing system according to claim 7, wherein 9. The information processing system according to claim 7, wherein the form recognition device has a storage unit for storing the form format data. 10. The form format data generating means includes means for generating registration table format data based on intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature amount extracting means. 10. The information processing system according to claim 7, wherein the means stores the registration table format together with a code for identifying the registration table format. 11. The form format data generating means includes means for generating search table format data based on intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature extraction means. The recognizing unit includes a similarity calculating unit that compares the search table format data with the registration table format data stored in the storage unit and calculates a similarity between the search table format data and the registration table format data. The information processing system according to claim 10, wherein: 12. An identification code of a registration table format data having a high similarity to the search table format data and the similarity are output as a form recognition result based on a calculation result obtained by the similarity calculation means. The information processing system according to claim 11, further comprising a form recognition result output unit. 13. An image feature amount extraction process for extracting a very small amount of form image data, a form format data creation process for generating form format data from line information obtained by the image feature amount extraction process, and the form format data And a form recognition process for performing form recognition based on the form. 14. An image input process for reading a form as image data, wherein the image feature amount extracting process extracts a feature amount of the form image data read by the image input process. Form recognition method described. 15. The method according to claim 13, further comprising a storage process for storing the form format data.
4. The form recognition method described in 4. 16. The form data creation processing includes a process of generating registration table format data from intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature extraction processing. 16. The form recognition method according to claim 13, wherein the processing stores the registration table format data together with a code for identifying the registration table format data. 17. The form format data generating process includes a process of generating search table format data based on intersection coordinate values of table ruled lines and cells in the form image data extracted by the image feature extraction process. The recognition process includes a similarity calculation process of comparing the search table format data with the registration table format data stored in the storage process to calculate a similarity between the search table format data and the registration table format data. 17. The form recognition method according to claim 16, wherein: 18. An identification code of registration table format data having a high similarity to the search table format data and the similarity are output as a form recognition result based on a calculation result obtained in the similarity calculation process. 18. The form recognition method according to claim 17, further comprising a form recognition result output process. 19. A storage medium storing a computer-readable program for executing a form recognition method for automatically recognizing a form, wherein the form recognition method includes an image feature for extracting a very small amount of form image data. An amount extraction step, a form format data generation step of generating form format data based on the line information obtained by the image feature amount extraction step, and a form recognition step of performing form recognition based on the form format data. Characteristic storage medium. 20. An image inputting step of reading a form as image data, wherein the image feature amount extracting step comprises:
20. The storage medium according to claim 19, wherein a feature amount of the form image data read in the image input step is extracted. 21. The storage medium according to claim 19, further comprising a storage step of storing said form format data. 22. The form data creation step,
Generating registration table format data from intersection coordinate values of table ruled lines and cells in the form image data extracted in the image feature extraction step; and the storage step identifies the registration table format data. 22. The storage medium according to claim 19, wherein the storage medium is stored together with a code to be executed. 23. The form data creation step,
A step of generating search table format data based on intersecting coordinate values of table ruled lines and cells in the form image data extracted in the image feature amount extraction step, wherein the form recognition step includes the search table format data and the storage step 23. The storage medium according to claim 22, further comprising a similarity calculation step of comparing the registration table format data stored in the search table format data and calculating the similarity between the search table format data and the registration table format data. 24. An identification code of registration table format data having a high similarity to the search table format data and the similarity are output as a form recognition result based on the calculation result obtained in the similarity calculation step. The storage medium according to claim 23, further comprising a form recognition result output step.