JPH08272884A - Form classifying and processing method and system - Google Patents

Abstract

PURPOSE: To speed up the character recognition processing of form which is entered by various kinds of characters and to smoothly cope with a recognition error. CONSTITUTION: In a form 1, the characters composed of only numerals are entered at each place. The characters composed of KANJI (Chinese character) are partially entered. In this case, the whole of the form is read by an image sensor 31 with relatively low resolution. The parts of KANJI are read by an image sensor 32 with high resolution. Thus, if a recognition processing is performed by each of a first recognition part 35 and a second recognition part 36, character recognitions can be exactly performed at a high speed and exactly by performing a signal processing with low resolution for a wide range and performing a signal processing with high resolution for a narrow specified range, respectively. A list printing is performed for a recognition impossible part as image data and a separate input is made possible.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a form classification method which receives a form such as a slip, reads the characters written therein, sorts into a large number of pockets, sorts and stores the form, and executes processes such as tabulation. Regarding the system.

[0002]

2. Description of the Related Art For example, in slip processing at supermarkets and convenience stores with stores nationwide, invoice processing for mail order, merchandise delivery and mail delivery processing, a large number of slips are classified, distributed and aggregated. Some work is required. Specifically, a form in which certain items are described is issued, the forms are collected, sorted by destination, and sent to the destination of the form. In the case of a slip or the like, in this case, the work of totaling the amount of money and the like described therein is involved. Such operations typically require large numbers of personnel and are extremely time consuming for processing. Therefore, in general, a large number of specialized personnel have been assigned to the slip processing work.
Further, in order to automate this kind of processing, a system in which a slip itself is eliminated and data is directly input and exchanged using a computer terminal is also widely used.

[0003]

The above-described conventional form processing system has the following problems to be solved. The present inventors have developed a device that recognizes characters on a form and sorts the form into a large number of pockets according to the contents thereof in order to automatically sort a large amount of forms. In such a device, high-speed and reliable recognition of characters written on a form and smooth processing for coping with recognition error are required. There is also a demand for a system that allows the operator to easily and surely check and process which form corresponds to which data and which form has a recognition error.

[0004]

The present invention adopts the following constitution in order to solve the above points. The method of the present invention reads the image of a form, recognizes the characters written on the form, and sorts and stores the form in the corresponding pocket based on the recognition result. Are read in parallel simultaneously by a plurality of different reading means, the corresponding recognition processing is executed, and the recognition results are combined to obtain the processing data for each sorted form.

The device of the present invention is an object to be recognized in a device that reads an image of a form, recognizes characters written on the form, and sorts and stores the form in a corresponding pocket based on the recognition result. A reading unit that reads characters under different resolutions, a character recognition unit that accepts image data obtained by each reading unit and executes recognition processing of the corresponding character, and a form that is sorted by combining the recognition results. And a synthesizing unit for obtaining processing data for each.

According to the above method, when any character to be recognized is unrecognizable, at least the corresponding part of the read image data is printed and a non-recognition list for separately keying is generated. Good to do.

At the same time when the image of the form is read, an identification code is printed on each form and an unrecognizable list is generated in association with this identification code.

[0008]

[Function] On the form, characters consisting only of numbers are described in various places. In addition, some characters are written in Kanji. In this case, the entire form is read by an image line sensor having a relatively low resolution. The kanji part is read by an image line sensor with high resolution. In this way, if the recognition processing is performed by the first recognition unit and the second recognition unit, respectively, the wide range can be accurately recognized by the low-resolution signal processing at high speed, and the specific narrow range can be accurately recognized by the high-resolution signal processing. The unrecognizable portion is printed as a list of image data, which can be input separately.

[0009]

The present invention will be described in detail below with reference to the embodiments shown in the drawings. FIG. 1 is a schematic explanatory diagram of the system of the present invention.
In the figure, in the system of the present invention, in order to recognize the characters written on the form 1, for example, two kinds of hardware for reading characters having different resolutions are used. The image line sensor 31 shown in the drawing is a reading device that covers the width W1 of the form 1 when the form 1 is conveyed rightward in the drawing. Further, the image line sensor 32 is a device that reads only the width W2 in order to read the Chinese character item written on the form 1. In this case, the image line sensor 31 has a relatively low resolution, for example, 170 dots per inch, that is, 170 d.
The image line sensor 32 has a resolution of about 340 dots per inch, that is, about 34 dpi.

These image line sensors 31, 32
Are output from the first reading unit 33 and the second reading unit 3, respectively.
Accepted by 4. The first reading unit 33 is a unit that receives the output of the image line sensor 31 and processes the image data, and sends the output to the first recognition unit 35 to cut out or recognize characters. The second reading unit 34 is a unit that receives the output of the image line sensor 32 and processes image data, and the data is sent to the second recognition unit 36 to perform character cutting and recognition processing. The results recognized by the first recognizing unit 35 and the second recognizing unit 36 are sent to the recognition result synthesizing unit 40, and are combined as data for one form. The result is stored in the processing data memory 41.

In this embodiment, as a result of recognition processing performed by the first recognition section 35 and the second recognition section 36, if the recognition is judged to be impossible, the read image data is the unrecognizable image data as it is. It is sent to the processing unit 42. The unrecognizable image data processing unit 42 is a unit that accepts such an image, associates it with the identification number of the form, and edits and controls the printout. In the unrecognizable list 43 thus obtained, the characters that cannot be recognized by the form are printed as an image as they are next to the identification number of the form. The operator is passed such an unrecognizable list, inputs an unrecognizable character using the keyboard 44, and sends it to the processing data memory 41. In this way, it is combined with the data already stored in the processing data memory 41 to complete the form-related data.

FIG. 2 shows a form processing explanatory diagram by the system of the present invention. The system of the present invention processes the form as a whole in such a procedure. First, the sender A1 fills in a large number of forms 1 with a destination, a billing amount, and other items. This form 1 is created for various destinations, and this form A is collectively sent from the source A1 to the sorting center. The sorting center is equipped with a form classification processing system as shown in FIG. Then, first, in step S1, the form 1 is read one by one, and the data is recognized. Then, in step S2, the forms are classified by destination. In this example, destinations B1 and B
The forms created by the sender A1 are sorted in the order of 2 and B3. These forms are finally sent to the destination as they are, but at this time, the billing amount is calculated for each destination. The billing amount is also written on the form and is recognized by the character recognition unit. Therefore, the following procedure is adopted to collect the recognition results and confirm the results.

First, in step S3, the destination B1
Divide the addressed form into 50 sheets. Then, 50 sheets are taken out and the reading results are totaled (step S3).
Step S4). These are automatically executed by the computer, and the subtotal of the charged amount is printed every 50 sheets (step S).
5).

On the other hand, separately from the 50 sheets, the image of the read character portion is converted into image data. Then, the necessary parts are once printed on the paper (step S6). That is, the billing amount and the like required for tabulating the form are printed side by side on paper as image data. This is totaled by the operator in step S7. The totaling by the operator may be actually performed using the 50 sheets themselves. However, for example, 50 sheets may include data related to privacy such as personal data. In such a case, a portion unnecessary for such tabulation is excluded, and only a necessary portion is once printed on a sheet and sent to tabulation. This enables privacy protection.

Further, the work of collecting a large number of forms at hand and compiling them is relatively complicated and error-prone. Therefore, if only necessary portions of the respective forms are edited and printed so as to be easily aggregated as image data, the processing becomes efficient. Due to these various advantages, in this example, the main part of the read image is once printed on paper, and the totalization is performed by the operator. Then, the totalization result and step S5
In step S8, the result of totaling is compared. If the two match, the result of recognizing and processing the characters of the form and the data conversion is the same as the result calculated by the operator by actually seeing the form, and it can be seen that there is no error in the calculation or recognition result. In this case, move to step S9,
The billing amount for 50 sheets is fixed. On the other hand, if the matching results do not match, the operator compares the read and recognized data with the image, checks the recognition error, etc., and corrects the data.

FIG. 3 is a schematic block diagram of the system of the present invention. As shown in the figure, the system of the present invention is
Is received in the hopper 2, the image of the form is read by the main body portion 3 while being conveyed by a conveying means (not shown), the characters written on the form are recognized, and sorting is performed based on the result. For this sorting, a large number of pockets 4 for storing the forms according to classification are provided. A console 5 is provided for controlling the apparatus, and a printer 6 is provided for printing out necessary data.

Some functional blocks provided on the main body 3 are shown on the lower side of this figure. As shown in the figure, the main body section 3 is provided with a reading section 11 for recognizing characters by reading the image of the form 1, and a character recognizing section 12.
Further, in the present invention, a bar code conversion section 13 and a bar code printing section 14 are provided to convert the result of such character recognition into a bar code and print the bar code in the margin of the form 1. . With this device, 1
It is possible to recognize the characters written on the form 1 by having a conveying unit that conveys the sheets one by one, and sort the forms into the corresponding pockets 4 according to the recognition result while conveying. However, the number of pockets 4 is specifically 36, for example, and when the form 1 is classified into about 300, it is necessary to set the form 1 in this system several times and repeat the classification and sorting.

In such a case, if character recognition or the like is performed each time, the recognition process may take a long time, and a recognition error is likely to occur. Therefore, in the present invention, a bar code that can be surely read and is less likely to cause a reading error is used. Therefore, for the second and subsequent classification / sorting processing, the barcode is read by the barcode reading unit 15 shown in the figure and the result is used. It should be noted that since the result of character recognition is used for the first time sorting and the output of the bar code reading unit is used for the second and subsequent times, both output signals are input to the sorting storage control unit 16. .

FIG. 4 shows a block diagram of the image reading portion of the form. As shown in this figure, the above-mentioned forms are thrown in from the hopper 2, and are sorted and stored in the pockets 4 through character recognition processing and the like. Four pockets are shown in the figure. The other pockets are omitted here. In order to recognize the characters written on the form, the reading unit 23, the reading unit 25, the skew correction image reading unit 26 and the like are provided in this embodiment. In addition, one half mirror 22 is arranged on the path for irradiating the form conveyed on the conveying path with light from a light source (not shown) and receiving the reflected light, as shown in the figure.

The reading unit 23 on the left side has 17 units per inch.
It has a resolution of 0 dots, that is, 170 dpi. Further, the half mirror 22 divides the reflected light of the form into two and sends them to the reading unit 25 and the skew correction image reading unit 26. The reading unit 25 has 340 dots per inch, that is, 340 dpi.
Read characters with resolution. This is mainly used for reading Chinese characters and the like written on a form. Furthermore,
The skew-corrected image reading unit 26 is provided to detect the inclination and the like during conveyance of the form. In this case, since resolution is not required so much, in this example, the skew-corrected image reading unit 26 is configured to read at a resolution of 170 dpi. Black characters and characters of other colors are processed by those with a higher recognition rate.

With the above configuration, the characters of the form conveyed on the conveying path from the hopper 2 are recognized, and various processes are executed. Specifically, it is necessary to recognize all the characters written on the form, convert them into barcodes, print the barcodes, and then sort and store them in pockets.

It should be noted that the form is conveyed with the surface facing down on the conveying path shown in FIG. The barcode printed on the back surface is read by the reading head 15A of the barcode reading unit 15. Further, the recognition result by each reading unit is converted into a barcode, and the barcode printing unit 1
Printing is performed by the print head 14A of No. 4. As described above, once the character recognition is completed, only the barcode reading unit 15 reads the classification of the form and controls the sorting process into the pockets 4.

FIG. 5 shows a more specific data read operation explanatory diagram. Letters, numbers, and other symbols are written on the form. As described above, in order to read and recognize the characters on the form that moves at high speed, a reading unit having a resolution corresponding to the characters written on the form is required. However, various characters such as numbers and Chinese characters are mixed in the form. For example, if the reading process is performed on the front side of the form using a high-resolution reading device, the character recognition operation cannot catch up with the supply of the form. On the contrary, if the reading is performed at a low resolution, the signal processing becomes extremely fast, but a recognition error or the like or the number of unrecognizable characters such as Chinese characters increases.

Therefore, as shown in the figure, for example, among the characters to be read on the form 1, simple characters such as numbers are read by the first reading unit, and complicated characters such as Chinese characters are read by the second. Take the method such as reading with the reading unit. It is assumed that the second reading unit performs reading with higher resolution than the first reading unit. In addition, in terms of the entire form, the area to be read by the first reading unit is increased, and the second reading unit reads the part that requires high resolution. As a result, the processing time of the signal is shortened, and moreover, it is possible to perform accurate recognition processing at a high resolution for those that require accurate recognition such as Chinese characters.

In order to perform a series of processes such as reading a form, recognizing characters, converting it into a bar code, and printing the bar code, it is necessary to extremely shorten the time until the recognition result is obtained. Therefore, on the assumption that the characters related to such a classification are numbers and codes that are relatively easy to recognize, such characters have a low resolution first.
It is read by the reading unit, immediately converted into a bar code, and printed. Then, it is sorted and stored in the hopper as it is. On the other hand, for the image data and the like read by the second reading unit having a high resolution, the form may be conveyed, and the character recognition may be completed before the reading of the next form is started. Therefore, even if the second reading unit executes the processing in parallel with the processing in the first reading unit and the result is obtained with a slight delay, it does not matter. By such a procedure, high-speed recognition and barcode printing are possible.

Returning to FIG. 1 again, in the present invention, a reading unit capable of reading at two or more kinds of resolutions is provided as described above, and it is easy to recognize even a relatively low-resolution process such as numbers. Performs signal processing at high speed. When the resolution is actually doubled, the signal processing speed becomes 1/4. Therefore, if the entire surface is read with high resolution and signal processing is performed, the signal processing may not catch up with the form processing. Therefore, the reading unit having a low resolution is used for reading except for the portion where the characters that require a high resolution for recognizing Chinese characters and the like are described. It should be noted that it is possible to provide three or more kinds of reading units or set several kinds of resolutions depending on the signal processing speed, the arrangement and size of the characters written on the form, and the like.

Further, a form that cannot be recognized may be rejected each time, and may be temporarily removed from the classification target. On the other hand, when a large number of forms arranged in a certain order are sorted into a certain classification, the order of the forms themselves may not be disturbed. That is, if the form is removed each time a recognition error occurs, it becomes inconvenient to restore the form.
Therefore, when many recognition errors occur, the forms are not sorted, the reading is finished as it is, and at the same time, the identification number is printed on the back face of the form. This identification number is a serial number for distinguishing each form. If the read image is printed in association with this identification number to obtain the unrecognizable list 43, the operator can input data by referring to only this unrecognizable list.

Moreover, the input data can be easily inserted into the corresponding portion in the processing data memory 41. This makes it possible to supplement or add recognition error data without taking out or referring to the form itself. If the number of recognition errors is very small, the sorting process or the like can be proceeded as it is. Further, the data related to the classification and sorting is surely recognized, and when the data of the other parts cannot be recognized, the processing can be naturally performed without disturbing the flow of the sorting processing. Of course, the identification code is not limited to the identification number given to each form, but may be data for identifying the form in the read characters of the form.

[0029]

In the form classification processing method and system of the present invention described above, the characters to be recognized are simultaneously read in parallel by a plurality of reading means having different resolutions, and the corresponding recognition processing is executed. Since the recognition results are combined and the processing data for each sorted form is obtained, character recognition is performed at a high resolution for the necessary parts by following the processing of the form that is read at high speed, and recognition errors, etc. Can be reduced. For unrecognizable characters, the corresponding part of the read image data is printed and an unrecognizable list for key input is created, so that the operator can easily supplement the necessary data with the list. You can Further, if the unrecognizable list is generated in correspondence with the identification code, the additionally supplemented data can be normally combined with the processing data stored in the memory without extracting or matching the forms. Also, since multiple unrecognizable lists are printed on one sheet, it becomes easier for the operator to input when keying in.
It also saves paper.

[Brief description of drawings]

FIG. 1 is a schematic explanatory diagram of a system of the present invention.

FIG. 2 is an explanatory diagram of form processing by the system of the present invention.

FIG. 3 is a schematic block diagram of the system of the present invention.

FIG. 4 is a block diagram of an image reading portion of a form.

FIG. 5 is a diagram illustrating a specific reading operation.

[Explanation of symbols]

 1 Form 31, 32 Image Line Sensor 33 First Reading Unit 34 Second Reading Unit 35 First Recognition Unit 36 Second Recognition Unit 40 Recognition Result Combining Unit 41 Processing Data Memory 42 Unrecognizable Image Data Processing Unit 43 Unrecognizable List

Claims (4)

[Claims] 1. When the image of a form is read, the characters written on the form are recognized, and when the form is sorted and stored in the corresponding pocket based on the recognition result, the characters to be recognized are A form classification processing method characterized in that a plurality of different reading means read simultaneously in parallel, the corresponding recognition processes are executed, and the recognition results are combined to obtain processing data for each sorted form. 2. A document in which an image of a form is read, the characters written on the form are recognized, and the form is sorted and stored in a corresponding pocket based on the recognition result. Reading unit that reads under different conditions, a character recognition unit that accepts image data obtained by each reading unit and executes recognition processing of the corresponding character, and a processing for each form sorted by combining the recognition results. A form classification processing system comprising: a synthesizing unit for obtaining data. 3. If any character to be recognized is unrecognizable, at least the corresponding part of the read image data is printed to generate an unrecognizable list for separate key input. The form classification processing method according to claim 1. 4. The form classification processing method according to claim 3, wherein an identification code is printed on each form simultaneously with the image reading of the form, and the unrecognizable list is generated in correspondence with the identification code.