JPS62138981A - Information recognizing device - Google Patents

Abstract

PURPOSE:To decrease the burden of an operator by synthesizing and outputting an input image and the recognizing result and confirming very easily the erroneous reading, rejection, etc., of the recognizing result. CONSTITUTION:The character pattern written on a slip 10 is image-formed on a photoelectric converter 11, through an AD converter 12, a pretreatment circuit 13, and a recognizing circuit 14, the recognizing result is sent to a recognizing result memory 16, outputted to a character generator 17 and converted to the image of the character, symbol, etc., having the prescribed font. The output of the AD converter 12 is sent to an image memory 15 to store one image of the slip 10 and stored. The image synthesized by an image synthesizing circuit 18 is printed out by a printer 19, and at an output image 10, the images 21A and 21B of an input slip 10 and recognizing result patterns 22A and 22B are synthesized to the prescribed position.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an information recognition device that optically reads image information and recognizes characters, symbols, etc. from the characteristics of the read image information.

[Prior art] Conventionally, in an information recognition device that optically reads and recognizes characters, figures, etc., inputting, reading, recognizing, etc. a book number, etc.
When performing a series of operations for outputting results, it is necessary to check whether there are misreadings, rejections, etc. in the recognition results.

As a confirmation method, the following processes were broadly classified.

Processing l.

The recognition results are sequentially displayed on the CRT screen, and if a rejection or the like occurs, the input character pattern is also displayed on the CRT to prompt the operator to make corrections.

Process 2. The recognition results are sequentially output by a line printer or the like, and after the recognition process is completed, an operator compares the input form with the printer output to check for misreading and rejection.

Processing 3: The input @ vote is displayed as an image on a CRT, and at the same time, the recognition result is also output on the same screen, so that the operator can constantly check for misreading or rejection.

[Problems to be solved by the invention] However, in the above-mentioned processes 1 to 8, for example, the process l
This is effective against rejection, but when misreading occurs, it is necessary to verify by comparing it with the input form. Also,
In process 1.3, if a plurality of forms are processed at once and corrections are made later, a huge amount of image memory is required. Further, in process 2, it is necessary to compare the printer output and the input form, which is very troublesome.

The present invention aims to eliminate the drawbacks of the above-mentioned prior art, and improves the output method of recognition results of an optical information recognition device.
The aim is to simplify the device and reduce operator labor.

As a means to solve this problem, for example, this embodiment includes an image memory that can store one page of the input image itself, and a recognition result memory that stores the recognition results. The apparatus includes a storage memory, a character generator that generates a pattern such as a character or a symbol from a code resulting from the recognition, an image memory, and an image synthesis circuit that synthesizes and outputs pattern signals from the character generator.

"Operation" In this configuration, one page of read image information is stored in the image memory, the image information is passed through a recognition circuit, the recognized result is stored in the recognition result storage memory, and the code of the recognition result is stored in the recognition result storage memory. The information converted into a pattern through a character generator and the information from the image information are combined and output.

[Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 6 is a diagram showing a general conventional configuration of an optical information recognition device. The character pattern written on form 10 is
An A-D converter 12 that converts the image onto the photoelectric converter 11 and converts it into an electric signal, and then converts it into a digital signal.
The recognition results are sequentially output through a preprocessing circuit 13 that performs character cutting, binarization, normalization, etc., and a recognition circuit 14 that performs character feature extraction and identification using well-known methods such as stroke extraction and contour tracking. be done.

In this embodiment, the recognition result is combined with the input form image and outputted by a printer such as a laser beam printer.

Figure 1 is a block diagram showing an example of the configuration.

10 to 14 are almost the same as figure 6, but A-D converter 1
In addition to the preprocessing circuit 13, the output of No. 2 is sent to an image memory 15 for storing the image of the form 10 - autumnal equinox, and is stored therein. Further, the recognition result output from the recognition circuit 14 is sent to the recognition result storage memory 16, and is also output to the character generator 17, where it is converted into an image of characters or symbols having a predetermined font. Next, the image patterns in the image memory 15 and the character generator 17 are appropriately combined in an image combining circuit 18, and the combined image is printed out on paper by a printer 19. Here, the recognition result storage memory 16 may be provided inside the information recognition device, or may be an external storage device such as an office computer. Reference numeral 100 is a control unit for controlling each of the circuits described above, and lot is a ROM in which an operating program of a flowchart shown in FIG. 3, which will be described later, is stored.

Further, FIG. 5 shows an example of the basic configuration of this embodiment.

In the figure, 51 is a CCD scanner for reading images, 52 is a laser beam printer, 53 is a character recognition circuit, 54 is a host computer, 55 is an automatic document feeder, 56 is a plurality of input forms, and 57 is an output image.

FIG. 2 shows an example of an output image obtained by the above configuration.

In the figure, 20 is the output image, and 21A and 21B are the input forms l.
This is the character information itself written on O. Output image @
20, the image of the input form 10 and its recognition result patterns 22A and 22B are combined at a predetermined position. Recognition result pattern 22A, 22Bt-1fj force image 2
Although it is completely arbitrary where on 0, it is placed directly below the character on which the recognition target character is written, an example is shown here.

Further, an example of the processing operation for obtaining such an output image will be described with reference to the flowchart in FIG. 3.

When inputting a form, first in step 31 place the form 56 on the automatic document feeder 55 of the CCD scanner 51,
In the following step S2, a reading start key on a keyboard or the like is input. Note that in this embodiment, the reading process is started by inputting from a keyboard, etc., but C
The CD scanner 51 may be provided with a process start switch or the like, and the process may be started by inputting the switch.

When the reading process starts, first the document feeder 5
5 sends one original document to the reading surface of the CCD scanner 51 at step S3. The CCD scanner 51 reads the image data on the document set on the reading surface in the following step S4, and the photoelectric converter 11. The signal is converted into a digital signal by the A-D converter 12 and sent to the preprocessing section 13.

The A-D converter 12 also sends the digital signal to the -H image memory 15, and stores it in the image memory 15 in step S5.

On the other hand, the recognition circuit 14 receives information output from the preprocessing section 13.
The character generator 17 executes recognition processing in step S6 and outputs the result to the character generator 17. At this time, the recognition result is also stored in the recognition result storage memory 16. In step S7, the character generator 17 converts the received recognized character code into a font pattern corresponding to the font pattern and outputs the font pattern to the image synthesis circuit 18. In step S8, the image synthesis circuit 18 synthesizes this font pattern with the image take from the image memory 15 and outputs it to the laser beam printer 19. In step S10, it is determined whether all the forms 56 sent to the automatic feeder 55 have been processed through this series of processes, and if the process has not been completed, the process returns to step S3.

Through the above-described processing, an output image 20 as shown in FIG. 2 is obtained.

Furthermore, as mentioned above, the patterns that are synthesized on the output image 20 are limited to only the recognition results; for example, as shown in Figure 4 of the Imperial Rescript, the date and time, or the complexity of inputting multiple forms in succession, can be avoided. In order to eliminate this problem, it is also possible to write a heading 43 such as a serial number or an identification name.

Furthermore, the printer itself may be of any type as long as it can output a single image, and is not limited to a laser beam printer. Therefore, this embodiment can be easily configured as long as there is a digital image scanner and a digital image printer. I can do it.

Furthermore, if the printer section is equipped with a function capable of two-color development, it is possible to output the input form image and the recognition result image in different colors.

As described above, according to this embodiment, since the input image pattern and the recognized recognition pattern are output on the same paper, it is extremely easy to check for erroneous recognition.

[Effects of the Invention] As described above, according to the present invention, since the input image and its recognition result are combined and output, it is extremely easy to check whether the recognition result is misread or rejected. It becomes possible to reduce the burden on the operator.

In addition, the recognition results and the input image are output on the same page,
Since it can be left as evidence, there are great advantages in terms of operation and management of the recognition device.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment according to the present invention; FIGS. 2 and 4 are diagrams showing an example of image output according to the embodiment; FIG. 3 is a flowchart of the operation according to the embodiment; FIG. 5 is a diagram showing the configuration of the entire apparatus when a laser scanning type digital copying machine is used, and FIG. 6 is a diagram showing the configuration of an example of a conventional information recognition apparatus. In the figure, 11... photoelectric converter, 12... A-D converter, 13... preprocessing circuit, 14... recognition circuit, 15...
...Image memory, 16...Recognition result storage memory, 17
Character generator, 18 Image synthesis circuit, 19 Laser beam printer, 20 Output image. Figure 4 Figure 5

Claims (2)

[Claims] (1) In an information recognition device that optically reads characters, symbols, etc. and outputs the recognized recognition results to a predetermined output device, a first storage means for storing image information of the characters, symbols, etc.; a second memory for storing the recognition results of the characters or symbols, etc.;
a storage means, a pattern generation means for generating a pattern corresponding to the recognition result of the character or symbol, etc., and a pattern generated by the pattern generation means and image information stored by the first storage means are synthesized. 1. An information recognition device comprising a synthesis means for (2) The scope of the claim characterized in that the compositing means synthesizes image information such as characters or symbols stored in the first storage means and the pattern generated by the pattern generating means so that at least they do not overlap. The information recognition device according to item 1.