JP4455830B2 - Form reading method and form reading apparatus for performing the reading method - Google Patents

Description

[0001]
[Industrial application fields]
The present invention provides a form reading method suitable for a form reading apparatus that optically recognizes characters written on a form such as a slip and outputs image data of the entire form, and a form reading apparatus suitable for the reading method. About.
[0002]
[Prior art]
In recent years, a form reading apparatus that optically reads and recognizes characters on a slip of a bank or the like uses not only the recognized characters but also the entire image data of the form and image data cut out for OCR for later visual confirmation and the like. It is hoped that the data will be output and stored. For this reason, the form reading device according to the prior art includes the entire image data obtained by reading the entire form including the format printed in the dropout color with the image scanner, and the image data of the reading area excluding the dropout color and the range not specified. Original binarized image data read by binarization is individually stored in the entire image memory and the OCR memory, and the original OCR image obtained by performing character recognition based on the binarized data stored in the OCR memory. The data and the entire image data of the form stored in the entire image memory are output.
[0003]
Further, the form reading device according to the prior art is for continuously reading a plurality of forms stacked in a predetermined direction on the hopper, but the direction of a part of the forms stacked on the hopper may be different, and this reverse direction In order to continuously read these forms, a rotation memory for rotating the image data read in the reverse direction is provided.
[0004]
For this reason, the form reading apparatus according to the prior art stores the aforementioned OCR memory for character recognition, a rotation memory for rotating the binarized data stored in the OCR memory, and the entire image data of the form. A total of four memories: an image memory for rotating the binary data stored in the image data, and a rotation memory for rotating the binarized data.
[0005]
Incidentally, as a document describing an optical character reader having a memory for rotating the image of the above-mentioned form, for example, JP-A- 2002-7960 can be cited.
[0006]
[Problems to be solved by the invention]
The above-described conventional form reading apparatus includes a reading process for reading image data of a color-printed slip and storing it in an OCR memory and an overall image memory, and a binarization stored in the OCR memory. A character recognition process for performing character recognition from data, and an output process for outputting and storing each image data stored in the image memory and the OCR memory to the outside. As described above, the above-mentioned data acquisition and output time is long. For this reason, the form reading apparatus according to the prior art has a problem that the time for writing and outputting to the memory becomes long and the reading time of the slip becomes redundant because the color image data has 24 times the amount of data compared to the binarized data. was there.
[0007]
An object of the present invention is to eliminate the inconveniences caused by the above-described prior art, and to provide a form reading apparatus capable of reading a plurality of color-printed slips at high speed.
[0008]
[Means for Solving the Problems]
To accomplish the above object, reads the color image data of a form, and a pair of OCR memory for storing the binarized image data obtained by binarizing the color image data, color image data in which the read the overall image memory pair for storing the binarized image stored in the OCR memory for data and output as well as storage and該全body image of the entire image memory of the color image data from the OCR memory A control unit that controls output from the memory, and the control unit performs character recognition based on the binarized image data stored in the OCR memory, and then converts the binarized image data and the color image data. A form reading method of a form reading apparatus for outputting from the OCR memory and the entire image memory,
The control unit is
A first step of storing the binary image data in one OCR memory and storing the color image data in one overall image memory;
A second step of performing character recognition of the binarized image data stored in the one OCR memory in the first step;
A third step of outputting the binarized image data stored in the one OCR memory and the color image data stored in the one overall memory;
During the execution of the second step, the binarized image data of the color image data of the next read form is stored in the other OCR memory and the color image data is stored in the other whole image memory. 4 steps,
A fifth step of performing character recognition of the binarized image data stored in the other OCR memory in the fourth step;
Executing the binarized image data stored in the other OCR memory and the sixth step of outputting the color image data stored in the other overall memory;
When the image direction of the binary image data and the image data stored in the one OCR memory and the entire image memory is different from a predetermined direction,
Stop the character recognition in the second step and the storage of the image data in the fourth step, rotate the binarized image data and the image data in different directions in the predetermined direction, and rotate the binarized image data and A seventh step of storing image data in the other OCR memory and the other overall image memory ;
The eighth step of performing character recognition of the rotated binary image data stored in the other OCR memory, the rotated binary image data stored in the other OCR memory, and the other whole image memory Performing a ninth step of outputting the stored rotated image data;
A first feature is that when the seventh step is executed, the tenth step of storing the binarized image data in the one OCR memory and the color image data in the one whole image memory is executed. .
[0009]
The invention further entirety reads color image data of a form, and a pair of OCR memory for storing the binarized image data obtained by binarizing the color image data, a pair of storing color image data of the read Image memory, storage of binary image data in OCR memory and output from OCR memory, and storage of color image data in overall image memory and control of output from overall image memory A controller that performs character recognition based on the binarized image data stored in the OCR memory, and then converts the binarized image data and color image data into the OCR memory and the entire A form reading device that outputs from a business image memory,
The control unit is
A first step of storing the binary image data in one OCR memory and storing the color image data in one overall image memory;
A second step of performing character recognition of the binarized image data stored in the one OCR memory in the first step;
A third step of outputting the binarized image data stored in the one OCR memory and the color image data stored in the one overall memory;
During the execution of the second step, the binarized image data of the color image data of the next read form is stored in the other OCR memory and the color image data is stored in the other whole image memory. 4 steps,
A fifth step of performing character recognition of the binarized image data stored in the other OCR memory in the fourth step;
Executing the binarized image data stored in the other OCR memory and the sixth step of outputting the color image data stored in the other overall memory;
When the image direction of the binary image data and the image data stored in the one OCR memory and the entire image memory is different from a predetermined direction,
Stop the character recognition in the second step and the storage of the image data in the fourth step, rotate the binarized image data and the image data in different directions in the predetermined direction, and rotate the binarized image data and A seventh step of storing image data in the other OCR memory and the other overall image memory ;
The eighth step of performing character recognition of the rotated binary image data stored in the other OCR memory, the rotated binary image data stored in the other OCR memory, and the other whole image memory Performing a ninth step of outputting the stored rotated image data;
A second feature is that when the seventh step is executed, the tenth step of storing the binarized image data in the one OCR memory and storing the color image data in the one whole image memory is executed. .
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a form reading apparatus to which a form reading method according to an embodiment of the present invention is applied will be described in detail with reference to the drawings. FIG. 1 is a diagram for explaining the overall configuration of a form reading apparatus according to the present embodiment, and FIG. 2 is a view for explaining a form reading procedure according to the present embodiment.
[0011]
As shown in FIG. 1, the form reading apparatus according to the present embodiment includes a control unit 1 that controls the entire apparatus, a keyboard 14 that is connected to the control unit 1 and inputs an operator's instructions, and various types of read A hard disk 12 for storing data, a display unit 13 for displaying correction candidates of read characters, and a form reading unit 2 for reading character data / whole image data / OCR data from a form, The form reading unit 2 includes a hopper 27 for stacking and storing a large number of forms 23 to be read, a conveyance path 22 for separating and conveying the sheets one by one from the hopper 27, and the conveyance path 22 The reading processing unit 21 for reading image data from the conveyed form, the printing unit 25 for printing predetermined characters or the like on the form conveyed on the conveyance path 22, and the reading and printing are performed. It comprises a delivery stacker 26 for reserving a form, and a reject stacker 28 for reserving a form that could not read by the operation of the switching unit 24.
[0012]
The form processing unit 21 includes an image sensor 21e for reading an image from a conveyed form, OCR memories 21a and 21c for storing the above-described binarized data for character recognition, and an entire image of the form. The entire image memory 21b and 21d for storing is provided.
[0013]
In the form reading apparatus configured as described above, the reading processing unit 21 sequentially conveys the form 23 stacked on the hopper 27 by the instruction of the control unit 1 through the conveyance path 22, and the reading processing unit 21 uses the OCR image data and the whole image for the form. Image data is stored in the OCR memories 21a and 21c and the entire image memories 21b and 21d, respectively, and the OCR image data and the entire image image data after character recognition are stored in the hard disk 12 or the optical disk 111. It will work.
[0014]
Next, a form reading operation based on the form reading method according to the present embodiment will be described with reference to FIG. FIG. 2 is a diagram for explaining the operation procedures of the OCR memory 21a, the entire image memory 21b, the OCR memory 21c, and the entire image memory 21d in time series. Note that “capture” shown in FIG. 2 is an agreement with “read” of image data. The numbers “111” to “666” in each of the memories 21a to 21d are examples of binarized data that is a character recognition target, and “ABC” is an example of an item example printed in dropout color. Is.
[0015]
First, the form reading apparatus according to the present embodiment stores the read OCR binarized data in the OCR memory 21a (capture a1) and the entire image data of the form when reading the first form. Store in the image memory 21b (take-in b1). The first read data is subjected to character recognition (recognition a1) in the OCR image memory 21a, and then the data is output from the memories 21a and 21b (outputs a1 and b1), respectively.
[0016]
The form reading apparatus according to the present embodiment reads the second form in time synchronization with the recognition a1 performing the first character recognition, and reads the read binary data for OCR into the OCR memory. The entire image data of the form is stored in the entire image memory 21d (capture d2). In other words, the form reading apparatus according to the present embodiment performs the character recognition process using the binarized data stored in the first OCR memory 21a, and the image of the second form in the OCR memory 21c and the entire image memory 21d. The form reading apparatus according to the present embodiment, which performs the data storage process in parallel in time, in the subsequent form reading operation, synchronizes with the output c2 / d2 of the second form from the memory 21a / 21b. The fetch e3 / f3 using the form memory 21a / 21b is performed. Similarly, the recognition of the third form e3 and the fetch of the fourth form g4 / h4 are performed in synchronization, and the third form The output e3 / f3 and the recognition of the fourth form g4 are operated synchronously, and the continuous form recognition, capture operation / output, and capture operation are performed by the OCR memory 21a / whole image memory. 21b and OCR media It performed using in parallel the memory 21d for Li 21c / the whole image.
[0017]
Therefore, since the form reading apparatus according to the present embodiment performs processing (reading-recognition-output process) on two forms in parallel in time, the form reading time can be reduced as a whole. In particular, in the form reading apparatus that reads and outputs the entire color form image data with a large amount of data, the processing time can be shortened because other processes are performed in parallel when reading and outputting the data.
[0018]
Further, as shown in FIG. 2, the form reading apparatus according to the present embodiment, for example, when the fifth form is reverse to the predetermined direction, the OCR data and the whole image data stored in the memories 21a and 21b with the capture i5 / j5. Is rotated and transferred to the memories 21c2 and 21d, the recognition i5-output i5 / j5 is performed using the memories 21c2 and 21d, and the next sixth form is processed using the memories 21a and 21b. You can continue.
[0019]
As described above, the form reading apparatus to which the form reading method according to the present embodiment is applied performs the recognition process, the output process, the reading process, and the output process in time when executing the image data reading process, the recognition process, and the output process. By performing in parallel, it is possible to shorten the reading time of a large number of forms, and it is possible to process the forms continuously even when the forms are read in the reverse direction.
[0020]
In the embodiment, character recognition using the binarized data stored in the first OCR memory 21a and storage of the second binarized data in the OCR memory 21c are timed. In parallel with each other, taking in e3 / f3 of the third form in synchronism with the output c2 / d2 of the second form, and recognizing e3 in the third sheet and the fourth sheet Although an example has been described in which the capture g4 / h4 is performed in synchronization, and the output e3 / f3 of the third sheet and the recognition g4 of the fourth sheet are performed in synchronization, the present invention is not limited to these. The present invention includes, without limitation, one using a pair of OCR and whole image memories and executing one of the processes read / recognized / output by both memories in parallel in time.
[0021]
【The invention's effect】
As described above, the form reading method and form reading apparatus according to the present invention are provided with at least one pair of OCR memory for storing binarized data and image memory for storing image data, and output image data from one overall image memory. The time for reading a large number of forms can be shortened by performing the output process to be performed and the next image data storing or recognizing process for the other image memory or OCR memory in parallel. A plurality of printed slips can be read at high speed.
[Brief description of the drawings]
FIG. 1 is a diagram for explaining an overall configuration of a form reading apparatus according to an embodiment of the present invention.
FIG. 2 is a view for explaining a form reading procedure according to the embodiment.
[Explanation of symbols]
1: control unit, 2: form reading unit, 23: form, 21: form reading processing unit, 21a to 21d: memory, 21e: image sensor, 27: hopper, 22: transport path, 22: form printing unit, 26: Discharge stacker, 28: reject stacker, 24: switching unit, 11: optical disk unit, 12: hard disk, 13: display unit, 14: keyboard.

Claims (2)

A pair of OCR memories that read color image data of a form and store binarized image data obtained by binarizing the color image data; a pair of overall image memories that store the read color image data; A control unit that controls storage of binary image data in an OCR memory, output from the OCR memory, and storage of the color image data in an overall image memory and output from the overall image memory. The control unit performs character recognition based on the binarized image data stored in the OCR memory, and then outputs the binarized image data and color image data from the OCR memory and the entire image memory. A form reading method for a form reading apparatus,
The control unit is
A first step of storing the binary image data in one OCR memory and storing the color image data in one overall image memory;
A second step of performing character recognition of the binarized image data stored in the one OCR memory in the first step;
A third step of outputting the binarized image data stored in the one OCR memory and the color image data stored in the one overall memory;
During the execution of the second step, the binarized image data of the color image data of the next read form is stored in the other OCR memory and the color image data is stored in the other whole image memory. 4 steps,
A fifth step of performing character recognition of the binarized image data stored in the other OCR memory in the fourth step;
Executing the binarized image data stored in the other OCR memory and the sixth step of outputting the color image data stored in the other overall memory;
When the image direction of the binary image data and the image data stored in the one OCR memory and the entire image memory is different from a predetermined direction,
Stop the character recognition in the second step and the storage of the image data in the fourth step, rotate the binarized image data and the image data in different directions in the predetermined direction, and rotate the binarized image data and A seventh step of storing image data in the other OCR memory and the other overall image memory ;
The eighth step of performing character recognition of the rotated binary image data stored in the other OCR memory, the rotated binary image data stored in the other OCR memory, and the other whole image memory Performing the ninth step of outputting the stored rotated image data;
A form reading method comprising: executing the tenth step of storing the binarized image data in the one OCR memory and executing the color image data in the one whole image memory when the seventh step is executed. . A pair of OCR memories that read color image data of a form and store binarized image data obtained by binarizing the color image data; a pair of overall image memories that store the read color image data; A control unit that controls storage of binary image data in an OCR memory, output from the OCR memory, and storage of the color image data in an overall image memory and output from the overall image memory. The control unit performs character recognition based on the binarized image data stored in the OCR memory, and then outputs the binarized image data and color image data from the OCR memory and the entire image memory. A form reading device,
The control unit is
A first step of storing the binary image data in one OCR memory and storing the color image data in one overall image memory;
A second step of performing character recognition of the binarized image data stored in the one OCR memory in the first step;
A third step of outputting the binarized image data stored in the one OCR memory and the color image data stored in the one overall memory;
During the execution of the second step, the binarized image data of the color image data of the next read form is stored in the other OCR memory and the color image data is stored in the other whole image memory. 4 steps,
A fifth step of performing character recognition of the binarized image data stored in the other OCR memory in the fourth step;
Executing the binarized image data stored in the other OCR memory and the sixth step of outputting the color image data stored in the other overall memory;
When the image direction of the binary image data and the image data stored in the one OCR memory and the entire image memory is different from a predetermined direction,
Stop the character recognition in the second step and the storage of the image data in the fourth step, rotate the binarized image data and the image data in different directions in the predetermined direction, and rotate the binarized image data and A seventh step of storing image data in the other OCR memory and the other overall image memory ;
The eighth step of performing character recognition of the rotated binary image data stored in the other OCR memory, the rotated binary image data stored in the other OCR memory, and the other whole image memory Performing the ninth step of outputting the stored rotated image data;
A form reading apparatus characterized in that when executing the seventh step, the tenth step of storing the binarized image data in the one OCR memory and storing the color image data in the one whole image memory is executed. .

Images