JP2860137B2 - Optical character reader - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to an OCR device (optical character reading device).

(Prior Art) In a conventional OCR apparatus, a form reading process is performed by a job which is a bundle of forms. That is, the operator inputs a form to the OCR device for each job. Then, the operator gives sheet size and read information by key input while viewing the job contents on the CRT screen. On the other hand, the OCR device performs character recognition on the documents stacked on the hopper based on the given read information, etc. Was.

However, OCR adopting such a job specification method
According to the apparatus, the operator has to repeat the operation of stacking the forms in the hopper for each job, inputting the job contents, and monitoring the end of the job, and automation of the form processing has been required.

(Problems to be Solved by the Invention) As described above, in the conventional OCR apparatus, the operator stacks the form on the hopper for each job and inputs the job contents.
Since job end monitoring is performed, there is a problem that a load is imposed on an operator, and automation of form processing has been required. The present invention has been made in order to solve such problems of the conventional OCR device, and the purpose is to identify each job by the OCR device even if the forms of different jobs are stacked together on the hopper. Accordingly, it is an object of the present invention to provide an OCR device capable of automatically performing character recognition on a form.

[Constitution of the Invention] (Means for Solving the Problems) In the present invention, photoelectric conversion means for photoelectrically converting an image of a form, character recognition means for performing character recognition on an electric signal obtained by the photoelectric conversion means, Extracting means for extracting job identification information and job end information described in a predetermined position; read information storage means for storing sheet size information and read information of a form corresponding to each of the plurality of job identification information; When the job identification information is extracted by the extraction means, the character recognition means is made to perform character recognition on the form input based on the corresponding sheet size information and read information in the storage means, and the sheet size information and the read Recognition processing control means for continuing character recognition based on information until the extraction means extracts job end information. We configured the histological character reader.

(Operation) According to the above configuration, the job identification information and the job end information described on the form are extracted by the extracting means, and the character recognition based on the sheet size information and the read information corresponding to the job identification information is performed based on the job end information. Continue running until extracted. This is repeated as long as the form is in the hopper, so that different jobs are continuously performed.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram of an OCR device according to one embodiment of the present invention. In FIG. 1, reference numeral 1 denotes a photoelectric conversion unit including a CCD image sensor or the like. An electric signal corresponding to the form image obtained by the photoelectric conversion unit 1 is stored in a memory (sheet buffer) 2. The recognizing unit 3 recognizes a character pattern of a signal stored in the memory 2 under the control of the control unit 6. Reference numeral 4 denotes a read information storage unit which stores sheet size information and read information respectively corresponding to a plurality of job identification information. Reference numeral 5 denotes an output unit such as a CRT display device. Reference numeral 6 denotes a control unit for controlling each unit, which is constituted by, for example, a microcomputer or the like, and performs recognition processing control based on a flowchart shown in FIG.

The OCR device configured as described above uses a form as shown in FIG. That is, a frame 9 in which a job number, which is job identification information, is printed at a predetermined position (the left end of the first row) of all the forms 7, 8 and job end information (for example, X) is printed at other predetermined positions. Is printed on the frame 10. A frame 11 is a text box.

Now, a case where the job of the document 7 ₁ to _7-m, are carried out and a job for document 8 ₁ to 8 _n to OCR apparatus shown in Figure 1. The operator job number "01" to the frame 9 of the form _71, as well as fill the job number "02" to the frame 9 of the form 8 _1, fill in the "X" frame 10 of the form 7 _m, 8 _n These are stacked and stacked on a hopper (not shown), and an OCR device is instructed to start reading.

Thus, the first sheet of the form ₇₁ is first conveyed, the memory 2 the image is photoelectrically converted by the photoelectric conversion unit 1
(See FIG. 3 300). At this time, the control unit 6
Detects whether the job has been completed by referring to its own job completion flag as shown in the flowchart of FIG. 3 (301). In this case branches to YES because it is form 7 of the _first sheet, resets the same flag, together with the order the reading of the job number to the recognition unit 3, should be recognized retrieve data from an area of the memory 2 throat Is given by the control unit 6. Therefore, the recognition unit 3 reads the pattern of the designated area from the memory 2, performs character recognition using a predetermined method such as a pattern matching method, and gives the result “01” to the control unit 6 (30).
2). Based on this, the control unit 6 reads the read information (information FC from which memory 2 the pattern should be read) corresponding to the job number “01” and the sheet size information from the read information storage unit 4 and recognizes the sheet size information. (303). Therefore, together with the recognition unit 3 holds the recognition result to read the rows of the form ₈₁ based on given information (304),
The control unit 6 is notified of the end of recognition of one sheet. The controller 6 determines that the hopper is “empty” by a sensor (not shown) provided on the hopper.
Is detected (305). Since not "empty" here, the process proceeds from the conveyance of the document 7 ₂ to step 301, branches to NO at the step 301 because the job completion flag is reset, the recognition by the recognition unit 3 continues. Thus,
Detection of job end information of the frame 10 and is recognized to form 7 _m is performed from the recognition unit 3 "X" is performed in step 304, the control unit 6 sets the job end flag.

Next, in step 301 the form 8 ₁ is transported, since the possible branches to YES, search and transfer sheet size information and reading information recognized and correspondingly described "02" to the frame 9 is performed (302, 303 ) The following forms 8 ₁ to 8 based on this information
The recognition of _n is performed. Then, set the end of the job flag is performed again by the recognition of the form 8 _n, and an empty hopper is detected (305), and branches to YES is ended in step 306. If the end of the job is not described in the final form 8 _n , the process branches from step 306 to NO. However, since there is no form, not shown, the control unit 6 detects the end and proceeds to the end processing. Will be. At this time,
The end is displayed on the CRT screen (output unit 5), and an operator intervention is performed.

[Effects of the Invention] As described above, according to the present invention, the job of character recognition can be switched one after another by the job identification information and the job end information entered in the form, so Can be automatically and continuously processed, and the load on the operator can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of the present invention, FIG. 2 is a diagram showing a form format used in an embodiment of the present invention, and FIG. 3 is a flowchart for explaining the operation of the embodiment of the present invention. . 1 ...... photoelectric conversion unit, 2 ...... memory 3 ...... recognition unit, 4 ...... read information storage unit 5 ...... Output section, 6 ...... controller _{7 1 ~7 m, 8 1 ~8} n ...... form 9 … Job identification information frame 10… Job end information frame 11… Character frame

Claims (1)

(57) [Claims] 1. A photoelectric conversion means for photoelectrically converting an image of a form, a character recognition means for performing character recognition on a pattern of an electric signal obtained by the photoelectric conversion means, a job identification information described at a predetermined position on the form, Extraction means for extracting job end information; read information storage means for storing sheet size information and read information of a form corresponding to each of a plurality of job identification information; and job extraction information extracted by the extraction means. Then, the character recognition unit is made to perform character recognition on the form input based on the corresponding sheet size information and read information in the storage unit, and the character recognition based on the sheet size information and the read information is performed by the extraction unit. An optical character reading device, comprising: a recognition processing control unit for causing a process to continue until job end information is extracted. .