JPS6115285A - Optical character reader - Google Patents

Abstract

PURPOSE:To secure adaptation between a character frame and a printed form by means of the ink of a color other than the drop-out colors, by forming the character frame and the characters within the frame into separate patterns and displaying the recognition unable characters in the images synthesized with the character frame. CONSTITUTION:The image of characters and a character frame held by a line buffer 13 are separated from each other by a detection/separation circuit 14 and divided into the character image and the character frame image. Then the separating and recognizing jobs are performed every character to reject the recognition unable characters. The dot pitch of the display color is changed for the character frame for discriminating from characters. Thus the read characters are displayed at the upper part of a screen of a CRT20. While a pattern of the rejected character is synthesized with the character frame and displayed at the lower part of the CRT screen.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an optical character reader (OCR), and more particularly to an improvement in the display of unrecognizable characters.

[Technical background of the invention]

OCR forms in which characters such as katakana, hiragana, numbers, and symbols are written to be read by OCR generally have character frames printed in advance to limit the size, pitch, etc. of the characters. This character frame is printed with a special ink called dropout color, so that when the OCR reads the characters, the character frame and the white part of the form are optically equivalent.

FIG. 6 of the accompanying drawings is an explanatory diagram of an OCR form. In addition, in the following description of the drawings, the same elements are indicated by the same reference numerals. Book W1 has a character frame 2 printed in dropout color, and within this is the characters <``1'' and `` as shown in the illustration.
The characters (numbers) ``3'', ``5'', and ``7'' are entered.Here, the characters 11111, ``13+1.1``7''
It is assumed that the character ``5'' has been correctly filled in, but the character ``5'' has not been correctly filled in (unrecognizable).

FIG. 7 is an explanatory diagram of the screen of the display device when the characters on the form shown in FIG. 6 are read by OCR. At the top left of screen 3 are the characters read and recognized from form 1, ``' 1 ''.
“°3゛” and “7” are displayed, and the unrecognized (reject) character 115. For I+, the symbol “?” is displayed. The symbol ``°'' is displayed to indicate that the character is a rejected character, and the pattern of the rejected character is also displayed as shown by the reference numeral 4.

(Problems with the background technology) In this way, conventionally, only the image of the characters written on the form was extracted, imported into OCR, and subjected to lx processing and reject processing, so the character frame was set in a dropout color. I had to print it out. For this reason, forms for OCR cannot be printed as easily as printing on general paper, and the printing unit cost is high.

[Object of the Invention] The present invention has been made to overcome the above-mentioned drawbacks of the prior art, and provides an optical character reader (OCR) that is applicable to documents whose character frames are printed with ink other than dropout colors. The purpose is to provide.

[Summary of the invention]

In order to achieve the above object, the present invention provides a photoelectric conversion means for photoelectrically converting a character frame and characters within the frame to output pattern information, a separating means for separating the character frame and characters from this pattern information, and a separating means for separating the character frame and characters from the pattern information. The present invention provides an optical character reading device that recognizes a character pattern that has been written and biases the display of a character pattern and a character frame pattern toward unrecognizable characters.

(Embodiment of the Invention) Hereinafter, an embodiment of the present invention will be described with reference to FIGS. 1 to 5 of the accompanying drawings. FIG. 1 is a block diagram of the same embodiment. Images of characters and character frames on the form are taken into the photoelectric conversion circuit 11 and converted into electrical signals, and then sent to the preprocessing circuit 1.
In step 2, the removal of miscellaneous parts, image shaping, etc. are performed. The line buffer 13 holds characters and character frames for each line as images, and the detection and extraction circuit 14 separates the image in the line buffer 13 into characters and character frames, and also extracts one character from each line of characters. To separate. The extracted and separated characters are normalized by a normalization circuit 15. In other words, the characters are aligned in size and moved to the center of the character frame. The recognition circuit 16 recognizes the normalized characters with reference to the standard patterns in the dictionary memory 17, and discriminates unrecognizable characters.

The editing circuit 18 separates the characters determined to be unrecognizable by the recognition circuit 16 and the character frames separated by the detection cutting circuit 14.
Editing is performed based on the input from the KEY board 19, and the results are displayed and output on the CRT 20. It can also be sent to the outside from a communication line via the interface circuit 21, or can be sent to the outside via a magnetic tape (MT) 22 or floppy disk (F).
D) Let 23 memorize it.

FIG. 2 is an explanatory diagram of a form used in the embodiment shown in FIG. The character frame 2 printed on the form 1 is not printed in drop color, but in normal ink (ink that is captured by OCR as an image).

FIG. 3 is an explanatory diagram of characters and character frame images of the form shown in FIG. 2. The device 1 shown in FIG. 1 performs processing such as detection, cropping and recognition on this image, and displays the image as shown in FIG. 4 on the CRT 20. Here, the character "6" is
This is a character that was rejected as being impossible.

FIG. 5 is an explanatory diagram of the operation of processing the image shown in FIG. 3 by the embodiment shown in FIG. As shown in block 101, the character and character frame images held in the line buffer 13 are separated by the detection cutout circuit 14 into a character image (block 102) and a character frame image (block 102).
block 103). This separation work can be performed relatively easily at t7 by utilizing the fact that the character frames are arranged at a predetermined size and pitch.

Next, character-by-character separation and [operation are performed, and unrecognized characters (in the example of Figure 5, number 3 [l character "6")]
is rejected (block 104).

Regarding the character frame, processing is performed such as making the dot pitch coarser and changing the display color in order to distinguish it from the characters (block 105).

In this way, the read characters are displayed at the top of the CRT screen 3, and the rejected character pattern and character frame are composited and displayed on the F section of the screen 3 (block 1
05).

In this way, when displaying a reject, the image of the character and the character frame can be combined and displayed, so that characters with similar shapes (for example, an apostrophe and a comma) can be easily distinguished based on their positional relationship with the character frame. Furthermore, the display method of characters and character frames can be changed so that character frames are also displayed so that it does not become difficult to distinguish between characters.

〔Effect of the invention〕

As described above, according to the present invention, a character frame and characters within the frame are captured as an image, the captured image is separated into a character frame pattern and a character pattern, and the characters are recognized.
Rejected characters are displayed as a composite image of the character frame, so the rejected character pattern can also be determined from the positional relationship with the character frame pattern, allowing operators to perform rejig 1 - character recognition and correction, etc. It is possible to obtain an optical character reading device that is easy to read, and is applicable to documents whose character frames are printed with ink other than dropout colors.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is an explanatory diagram of a junior member used in the embodiment shown in FIG. 1, FIG. 3 is an explanatory diagram of an image of the form shown in FIG. 2, and FIG. 4 is an explanatory diagram of the CRT screen when the light truck shown in FIG. 2 is processed by the embodiment shown in FIG. 1, FIG. 5 is an explanatory diagram of the image processing operation of the embodiment shown in FIG. Figure 6 is an explanatory diagram of a conventional OCR form, Figure 7
The figure is an explanatory diagram of a display screen when the form shown in FIG. 6 is processed by a conventional device. 1...Form, 2...Character frame, 3...Screen. Station Ward Ward [F] Toren Tan

Claims (1)

[Claims] A photoelectric conversion means for photoelectrically converting a character frame preprinted on a form and characters recorded in the character frame to output pattern information; Separation means for separating into patterns; Character recognition means for recognizing the separated character patterns; and Displaying, for characters that the character recognition means cannot recognize, the character pattern and a character frame pattern corresponding to the character pattern. What is claimed is: 1. An optical character reading device comprising display means for