JP2812392B2 - Character processing apparatus and method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character processing apparatus and method, for example, a character processing apparatus that performs character recognition based on image data and stores a resulting character code in a memory. And methods.

2. Description of the Related Art Conventionally, in a device of this type, when an erroneous entry is made on a mark sheet used as a character recognition sheet, an erroneous entry on the sheet is erased with an eraser or changed to a figure in which character recognition is not possible. And make corrections.

[Problem to be Solved by the Invention] However, the above conventional example has the following disadvantages.

(1) When erasing with an eraser, dirt will adhere to the sheet.

(2) When erasing with an eraser, there is a possibility that various marks necessary for recognition are blurred.

(3) By the above (1) and (2), erroneous character recognition is likely to occur.

(4) If characters are entered with a felt-tip pen,
It takes time to rewrite mistakes.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-described drawbacks of the related art, and has as its object to provide a character processing apparatus and method capable of easily correcting erroneously entered characters.

Means for Solving the Problems A character processing method according to the present invention for achieving the above object is a character processing method for recognizing characters of input image data and sequentially storing character codes of the recognition result in a memory. And cutting out one character image data from the input image data, determining whether a character code of a recognition result of the cut out one character image data is discard information, and a result of the determination, If it is determined that the character code is not the discard information, the character code of the recognition result is stored in the memory.If the result of the determination is that the character code of the recognition result is the discard information, Characters that are a predetermined number of characters before the character of the recognition result determined as the discard information are discarded from the memory.

Further, a character processing apparatus according to the present invention for achieving the above object has the following configuration. That is, a character processing device that recognizes characters of input image data and sequentially stores a character code of a recognition result in a memory, wherein a cutout unit that cuts out one character image data from the input image data, Determining means for determining whether the character code of the recognition result of the cut-out one-character image data is discard information; and, as a result of the determination by the determining means, it is determined that the character code of the recognition result is not discard information. In the case, the storage control means for controlling the character code of the recognition result to be stored in the memory; and, as a result of the determination by the determination means, when the character code of the recognition result is determined to be discard information, Discarding control means for discarding a predetermined number of characters before the character of the recognition result determined as the discard information from the memory. Obtain.

[Operation] According to the above configuration, when it is determined that the character code of the recognition result is not the discard information, the character code of the recognition result is stored in the memory, and the character code of the recognition result is determined to be the discard information. In this case, the character that is a predetermined number before the character of the recognition result is discarded from the memory. As described above, switching between “processing for storing the character code” and “processing for discarding a character preceding the character by a predetermined number from the memory” is performed depending on whether the character code of the recognition result is the discard information. There is no need to provide a special area for writing information, and processing for identifying discarded information can be performed simply and reliably.

Embodiment A preferred embodiment according to the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an embodiment of a character recognition device according to the present invention. In FIG. 1, reference numeral 1 denotes a character recognition device of the present embodiment. Reference numeral 2 denotes input of image data of the mark sheet by a facsimile device or an image scanner of an external device which optically reads information such as characters written on the mark sheet, and outputs a result of character recognition based on the input image data to an external device. 2 shows an input / output circuit that outputs to an apparatus. Reference numeral 3 denotes a CPU that controls the entire apparatus. Reference numeral 4 denotes a ROM that stores a control program for operating the CPU 1, an error processing program, and a program according to a flowchart shown in FIG. 3, which will be described later. Reference numeral 5 denotes a work area for various programs stored in the ROM 4. Used as a temporary save area during error processing
Shows RAM. Reference numeral 6 denotes a memory for storing image data input from an external device via the input / output circuit 2 and a character code of a recognition result. Reference numeral 7 denotes a character cutout mark 101, 102 for each line on a mark sheet, which will be described later.
A character extraction circuit for extracting characters for each character, that is, for each character frame 103 is shown. 8 is a character extracting circuit 7
2 shows a character recognition circuit that recognizes what character pattern the character pattern in the character frame 103 cut out by and outputs the recognition result as a character code. The character recognition circuit 8 has a dictionary unit (not shown) in which standard patterns for matching with character patterns are recorded. Reference numeral 9 denotes a discard mark determination circuit for determining whether or not the mark is a previous character discard mark 104 according to the present embodiment, which will be described later, based on the result of recognition by the character recognition circuit 8. When the previous character discard mark 104 is detected, the discard mark determination circuit 9 functions to discard the immediately preceding recognition result on the memory 6. Reference numeral 10 denotes a bus line for transmitting address signals, data, and control signals in the apparatus.

FIG. 2 is a diagram showing an example of the mark sheet of the present embodiment. In FIG. 2, reference numeral 100 denotes a mark sheet of the present embodiment, and reference numerals 101 and 102 denote character cutout marks which are printed in advance to indicate the start and end of character cutout for each line. Reference numeral 103 denotes a character frame for designating a place where a character is to be written by a writer within a range sandwiched between the character cutout marks 101 and 102. 104 is character frame 1
A previous character discard mark for instructing discarding of a character entered in the immediately preceding character frame among characters entered in 03 is shown. In this embodiment, an asterisk mark “*” is used as the previous character discard mark 104. For example, in the mark sheet 100, the previous character discard mark 104 is written in the third character frame 103 on the second line. Therefore, the numeral character “6” written in the second character box 103 on the same line is discarded.

 Next, a character recognition method according to the present embodiment will be described.

FIG. 3 is a flowchart for explaining the operation of the CPU 3 of the character recognition device 1 of this embodiment.

First, the image data of the mark sheet is input from an external device via the input / output circuit 2 and stored in the memory 6 for one page (step S1). Then, in the memory 6, the character cutout marks 101 and 102 on the first line are detected by the character cutout circuit 7 (step S2). Next, one character (image data) in the leading character frame 103 existing between the character cutout marks 101 and 102 detected in step S2 is cut out by the character cutout circuit 7 (step S3). The cut-out image data for one character is collated with a standard pattern by a character recognition circuit 8 by a known technique, and the character is recognized (step S4). In this way, the character code of the recognized character output from the character recognition circuit 8 is collated with the character code of the previous character discard mark 104 by the discard mark detection circuit 9 (step S5). If the discrimination mark detection circuit 9 determines that the recognized character is not the previous character discard mark 104, the character code of the corresponding recognized character is stored in the memory 6 (step S7). Here, since the character recognition of the first line has not been completed yet, the process proceeds to step S3 again, and the character recognition of the character pattern in the second character frame 103 and the determination as to whether or not the previous character discard mark 104 is made. For example, in the case of the mark sheet 100 shown in FIG. 2, since the previous character discard mark 104 is not entered on the first line, steps S3 to S8 are repeated in this processing. When the first line is completed, the next line is still continued, so that the character cutout marks 101 and 102 of the next line are detected (step S).
10, step S13). Then, the character recognition of the second line and the determination as to whether or not it is the preceding character discard mark 104 are continued. This second
In the line, when the character recognition of the character pattern in the third character frame 103 and the determination as to whether or not it is the previous character discard mark 104 are performed in steps S3 to S6, the recognized character is determined as the previous character discard mark 104, Thereafter, the previous character is discarded. Therefore, in order to discard the recognized character "6" immediately before the recognized character determined as the preceding character discard mark 104, the character code of the recognized character "6" in the memory 6 is first cleared (step S12). Then, the address on the memory 6 in which the character code of the recognition character "6" is cleared is set as the storage location of the next recognition character, that is, the character code of the numeral character "3" following the asterisk mark "*" (step S).
3) In this way, the flow returns to step S3 again, and character recognition of the next character (in this case, the numeral character "3" corresponds) and determination of whether or not the character is the previous character discard mark 104 are performed. Here, the process proceeds to step S7 without being determined as the previous character discard mark 104, and the character code of the numeral character "3" is changed to the above-described step S1.
It is stored in the address of the memory 6 set in step 3. Thereafter, the processing is repeated until the character recognition for one page is completed, and if the character recognition for the one page can be output to an external device at the output destination connected to the input / output circuit 2, the recognition result for one page in the memory 6 is obtained. Is output (step S14).

As described above, according to the present embodiment, even if an erroneous entry is made on a mark sheet, the previous character is securely discarded by merely writing the discard mark of the previous character in the next character frame. It performs accurate character recognition by correction.

According to this embodiment, the preceding character discard mark is an asterisk mark “*”, but the present invention is not limited to this, and may be a mark such as “□”.

Further, according to the present embodiment, the recognized character that was previously recognized by the previous character discard mark is discarded. However, the present invention is not limited to this, and up to two or three characters can be discarded. The previous character discard mark may be set so as to discard up to the previous recognized character.

[Effects of the Invention] As described above, according to the present invention, a predetermined number of characters from a character of a recognition result determined to be the discard information are determined by whether or not the character code of the recognition result is determined to be the discard information. It switches between discarding the previous character from memory or storing the character code of the recognition result in memory,
It is not necessary to provide a special area for writing the discard information, and the effect that the process for identifying the discard information can be easily and reliably performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a character recognition device according to the present invention, FIG. 2 is a diagram showing an example of a mark sheet of this embodiment, and FIG. 3 is a block diagram of a CPU 3 of the character recognition device 1 of this embodiment. It is a flowchart explaining an operation. In the figure, 1 ... character recognition device, 2 ... input / output circuit, 3 ...
CPU, 4 ROM, 5 RAM, 6 memory, 7 character extraction circuit, 8 character recognition circuit, 9 discard mark determination circuit, 10 bus line, 100 mark sheet, 101, 102 …… Character cutout mark, 103 …… Character frame,
104: This is the previous character discard mark.

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06K 9/62

Claims (2)

(57) [Claims] 1. Recognizing characters of input image data,
A character processing method for sequentially storing a character code of a recognition result in a memory, wherein one character image data is cut out from the input image data, and a character code of a recognition result of the cut one character image data is discard information. And determining whether or not the character code of the recognition result is not discard information, storing the character code of the recognition result in the memory, and determining the character code of the recognition result. A character processing method comprising: when a character code is determined to be discard information, discarding, from the memory, a character a predetermined number of times earlier than a character of a recognition result determined to be the discard information. 2. Recognizing characters of input image data,
What is claimed is: 1. A character processing device for sequentially storing a character code of a recognition result in a memory, comprising: a cutout unit that cuts out one character image data from the input image data; and a character that is a recognition result of the one character image data cut out by the cutout unit. Determining means for determining whether or not the code is discard information; and as a result of the determination by the determining means, when it is determined that the character code of the recognition result is not discard information, the character code of the recognition result is stored in the memory. A storage control unit configured to control to store the character code of the recognition result as the discard information as a result of the determination by the determination unit; And a discarding control unit for discarding a few characters from the memory.