JPS6297081A - Character recognizer - Google Patents

Abstract

PURPOSE:To deliver a misrecognized part of character recognizing results after correcting it accurately and to reduce the misrecognizing factor of characters, by providing a resemblance calculating circuit and an inspection processor consisting of a memory and a microprocessor to perform recognition of words. CONSTITUTION:A character observing part 4 of a character recognizer 1 recognizes the characters written on a form for each line and delivers 6 them to a microprocessor 20 of an inspection processor 10 in the form of a character code string. The processor 20 delivers two character codes to a resemblance calculating circuit 30. Then the circuit 30 reads two standard patterns corresponding to said character codes out of a standard pattern storage 5 and sends them back to the processor 20 after calculation of their resemblance. Then the processor 20 performs first the recognition of words (key items) and designates the types of characters of the fixed items via a memory 11 to check the character codes. For the character codes corresponding to the key items, the character code string of a word dictionary, i.e., the work recognizing result is substituted for the character code string of the key items. Thus the misrecognition is corrected accurately and the misrecognizing facto of characters is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a character recognition device having a function of correcting misread characters, which is suitable for cases where there are many character categories such as kanji.

[Prior Art] Conventionally, most of the processing of various applications in government offices and the like has been done manually. These application forms are usually written in a mixture of kanji and kana, and if the application processing task were to be automated, the input unit would need a recognition device for Japanese characters, including kanji. Currently, at the laboratory level, we have successfully conducted a principle experiment of a printed kanji recognition device with reading accuracy that is satisfactory for practical use (for example, see Journal of the Institute of Electronics and Communication Engineers, Vol. 58-D, No. 2, p. 94).5 Considering that most of the various application forms mentioned above are relatively high-quality type-printed documents in Japanese type, it can be said that the environmental conditions for using the printed kanji recognition device in the above-mentioned application processing work are in place.

However, if a printed kanji recognition device is to be put into practical use, a fairly high level of recognition accuracy is required due to the nature of the application process. On the other hand, considering that kanji has an extremely large number of characters, and even if the print quality is relatively good, it is possible that an application form of relatively poor quality may be input, so the reading accuracy is quite sufficient. It can not be said.

However, it is conceivable that the rate of misrecognition can be significantly reduced by testing whether or not the recognition results are correct.

Conventionally, the above idea was carried out as follows. Character recognition devices that target numbers often handle monetary amounts, so for example, the amount of each item and the total amount are written on a form, and the recognition device can write the total amount of recognition results for each item, A method is used to detect errors by comparing the total recognition results. In addition, in character recognition devices that target English letters, N-gra
Although the method called m was used, other verification methods can be considered.

However, the above conventional method cannot be directly applied to a character recognition device for Chinese characters.

The reason for this is that in the case of kanji, the number of character types is 2000 to 4000 compared to alphanumeric characters (50 characters at most), and for example, the storage capacity of an N-gram table becomes enormous, making it impossible to realize it as it is. It becomes possible.

[Object of the Invention] Accordingly, the object of the present invention is to provide a means for correcting reading results using word information as a method suitable for cases where there are many character types;
The objective is to reduce the overall misrecognition rate.

[General description of the invention] In order to achieve the above object, the present invention is characterized in that the reading result is compared with the word in the word dictionary, and if there is a mismatch, the character code in the word is replaced and corrected. There is.

[Embodiments of the invention] For example, consider an application with the following documents:

(Example) Type of application Registration application Purpose of registration All transfers Cause February 2, 1971 Name of purchase and sale right holder Name of applicant Atabe Office 1-1 Kunitachi-shi, Tokyo Equity 1/3rd of obligor Name: Otsuyo Fukibe Address: 2-2, Tachikawa-shi, Tokyo Application date: March 3, 1978 The outline of the principle of the device of the present invention will be explained using the flowchart shown in FIG. First, at 201 and 202, characters on a form are photoelectrically converted, cut out within a certain frame, recognized line by line, and the recognition result is output for one line in the form of a character code. The recognition unit continues the above operation until all characters on the form are recognized. Everything up to this point is the same as the conventional character recognition device. Next, the recognition result verification unit extracts the character sequence of the recognition result corresponding to the key item (character string printed in a field of a predetermined length on the left side of one line), and creates a dictionary in which all key items are stored. It is recognized in step 203 to which key item this character sequence corresponds. This is called key item word recognition. Note that character recognition may be incorrect, so the word recognition method described above needs to be devised. The method will be described later.

Once the number of the key item is known, the character types that can appear in the fixed items following this key item can be limited, and the character type is specified in step 204. Therefore, in the character code string resulting from character recognition, the character code corresponding to the fixed item is checked to see if it is included in the above-mentioned permissible character types in step 205. At this time, if it is found that it is not included, either the result of character recognition is incorrect or the characters on the form are misspelled. Therefore, in this case, the above information is output along with the recognition result. For example, reverse the sign of the character code. As a result of the test, if the character type is included in the allowable character types, it is assumed to be read correctly and the character code is output as is.

Continue the above operations until there are no more characters on the form.

Next, a method for recognizing a character sequence resulting from recognition as a word, which is the main point of the present invention, will be explained. Generally, in order to recognize words, it is sufficient to prepare a word dictionary (a table consisting of character code strings that make up each word) and check which dictionary item the input character sequence matches. However, in reality, not all input character sequences are read correctly, so it may not be possible to find a perfect match with any dictionary entry6. It is necessary to perform word recognition by determining the distance or equivalently the degree of similarity (to be defined later) between the input character sequence and each dictionary entry.

For example, "Application date" may be obtained as a result of reading "Application date", but there is clearly no dictionary entry for "Application date"6. In the above example, the similarity between "Monkey" and "A" is required. However, such combinations of two characters are 400,000 combinations when the number of characters to be read is 2000 characters, and it is impossible to store them. therefore,
In the device of the present invention, when the similarity between different characters (in the above example, "K" and "Mon" is required, the similarity between the corresponding standard patterns in the recognition device is calculated and that value is used. .The degree of similarity between the same characters is always 1.

Here, the degree of similarity is a numerical value ranging from O to 1, which is defined between two character patterns, is easily calculated by a dedicated calculation circuit, and is well known, so its explanation will be omitted here.

The word recognition algorithm using the above method will be explained using the flowchart shown in FIG. First, each dictionary entry has the number of characters Nk that makes up the word, and the character code string Wk = (wi(k)
It is expressed as l i=L L ”'p Nk). Let K be the total number of dictionary items. In the above, k is the item number (
word number) and takes values from 1 to K. Further, the character sequence (character code sequence) of the character recognition result input to the word recognition unit is expressed as 5=(s il i=1.2. . . . =, N). The degree of similarity between character series S and wk is expressed as ρ.

Figure 2 shows the structure of a dictionary necessary for word recognition.

1 The first word 501 (number D) of the dictionary holds the number K of key items, then the address A 1 z A 23, which stores the character code string of each item, the word 50 which stores AN.
2 follows. Next follows a word that stores the character code string of each key item. For example, the A1 address 503 stores N1 (length and number of characters) that constitute the word of item number 1, and the following N1 word 504 stores each character code.

FIG. 3 illustrates a character code string that is a target of word recognition. The character code string is stored in the working area of memory,
Consists of N words.

In FIG. 4, word recognition is performed as follows. First, initialization is performed in steps 101 and 102.

In step 103, it is determined whether the word length matches the input character sequence length, and if they do not match, the similarity ρ is left as O and the next word is examined. When the word lengths match, the degree of similarity ρk is calculated in steps 105 to 112.

Initialization is performed in 104, and in 105 it is checked whether the i-th character code wx(k) of the th item in the dictionary matches the i-th character code sl of the input character series, and if they do not match, , 106 adds 1 to ρk, and if they do not match, it is checked in 107 whether it could not be determined. 5
When b=Q, it indicates that it is impossible to determine, in this case 106 is executed, and when s1≠0, in 108, the standard pattern in the recognition device is used to compare the standard pattern of Wi(k) and s
Calculate the similarity of standard patterns of l and add it to ρk. It is determined in 109 whether the value obtained by dividing ρk by the number of characters i up to that point exceeds the threshold value ε, and if it does not, the item is excluded from the candidates in 113. If the number exceeds the number, the process advances to the next character, and when the processing in steps 105 to 111 is completed for all characters, in step 112, the degree of similarity between the character series is divided by the number of characters N to normalize.

When it is detected in 115 that all dictionary items have been processed, the total similarity (ρkl k
``1 * 2s..., K) maximum value ρ and next largest value ρ
is calculated, and the absolute threshold value δ and ρ1 are compared and 117. Furthermore, whether or not there is a sufficient difference between ρ1 and ρ2 is tested using a relative threshold value γ, and when it is 1, output the word number k that gives ρ at 119. If it is not sufficient, it is impossible to judge. Output at 120.

Next, a method for specifying character types that can appear in fixed items following a key item will be explained. In the present invention, a flag table as shown in FIG. 5 and a bit number conversion table as shown in FIG. 6 are prepared. When the word recognition result of the key item is k, first refer to the bit number conversion table and bit position number b indicating which bit of the flag table is used.
Find (k). Next, extract the contents of the flag table for a given character, and if the value of the b(k) bit item is 1, the same character is allowed as a character type following the same key item, and if it is 0, it is not allowed. I understand.

Therefore, using this result, the recognition result can be verified as described in the explanation of the principle.6 Hereinafter, the present invention will be described in detail with reference to examples.

FIG. 7 is a block diagram of an embodiment of the apparatus of the present invention. Hereinafter, an embodiment will be described according to the same figure.

In the figure, 1 is a conventional character recognition device, 3 is a character observation section that observes unknown patterns, 4 is a character recognition processing device, and 5 is a character recognition device.
is a standard pattern storage device.

Since the above-mentioned parts are well known, they will not be described in detail here.

The output 6 of the recognition processing device 4 is the result of recognizing characters on a form line by line, and is transferred in the form of a character code string. Here, when the character code is 0, it means that the character is unrecognizable.

The test processing device 10 includes a memory 11 and a similarity calculation circuit 30.
and a microprocessor 2o. circuit 30
receives two character codes from the microprocessor 20 and generates two standard patterns corresponding to the same character codes into 5
Then, the similarity between the same standard patterns is calculated, and the resulting similarity is transferred to 20. The circuit 3o is used when executing the process 108 in FIG.

The memory 11 includes a section 12 for storing the flag table shown in FIG. 5, a section 13 for storing the bit number conversion table shown in FIG. 6, and a section 1 for storing the key item dictionary shown in FIG.
4 and a work area 15.

The microprocessor 20 performs word recognition (key item recognition) using the algorithm explained in FIG. 4 according to the microprogram contained in the microprocessor 2o.

The character type of the fixed item is specified using step 12, and the character code that is the recognition result of the fixed item is verified.

Next, the flow of processing as a single character recognition device will be explained.

The character pattern applied on the form is charged and converted by 3, cut out within a certain frame, and transferred to 4. In step 4, the degree of similarity between the unknown pattern sent from step 3 and each standard pattern in step 5 is calculated, and the coats of characters that give the maximum degree of similarity are summarized for one line and output as a character code string on output line 6. Output 6 However, here, 4 tests whether the maximum similarity is greater than a predetermined threshold, and if it does not reach the threshold, the output code k is set to 0.

The microprocessor 20 in the verification processing device lo receives the character code string of the recognition result for each line through 6 and stores it in the memory 15. First, a character code sequence corresponding to a key item is extracted from a character sequence for one line (blanks are also given one character code), and the process moves on to word recognition. An example of a character code string for one line is shown in FIG. One line consists of 25 characters, the first 8 characters 801 correspond to key items, and the latter 17 characters 802 correspond to fixed items. The character code 9999 means blank and is a non-blank character code in 1eJ801 (in Figure 8, S'l+S2
+..., s6) is a character code string as a result of recognizing the characters of the key items.

Word recognition is performed by a microprogram according to the algorithm shown in FIG. However, in the same algorithm, the process 108 in FIG. 4 is performed by a similarity calculation circuit. In other words, 20 is composed of two character codes Si and 1 (
1st character code wi(k) (4th
(see figure) is transferred to 30, and a similarity calculation command is issued to 30.

30 receives the same command and converts 2 corresponding to Si and wi(k).
The standard patterns 5 are read out from 5, the similarity ρ between the same standard patterns is calculated, and the result is returned to 20. The above is the process 203 in FIG.

When the microprogram word recognition is completed, the process moves to verification processing. First, processing 204 in FIG. 1 is performed.

First, when the key item number as a result of key item recognition is known, the bit number conversion table in the memory 13 is checked and bit number b of the flag table that specifies the character type of the fixed item following the key item is determined. obtain. Next, verification processing 2 of recognition results for fixed items
Do 05. 2.0 is the character code 8 corresponding to the fixed item in the recognition result character code string (Fig. 8) in the memory 15.
02 one by one and the flag table in memory 12 (
Check the b-th bit of the flag corresponding to each character code (see FIG. 5). When the same bit is 1, it is an acceptable character type, so nothing is done at that time, but when it is 0, it is an unacceptable character type, so the sign of the character code in 802 that gave the same result is reversed. let For example, if the recognition result character code with a fixed item is 500, and the character is not allowed as a result of the verification, the same code is inverted and set to -500.

Here, if the character code of the fixed item recognition result is a negative sign at the stage when it is sent from 4, the verification process for the same character code is not performed.

In addition, the character code string corresponding to the key item 1 (initially, the character code string of the dictionary of word recognition results is substituted into the character code string of the key item shown in FIG. 8. For example, if the character recognition result 801 is If the word recognition result is a key item number corresponding to "Application mortar" even if it is "Application mortar", the microprocessor 2o stores the character code string corresponding to "Application mortar" instead of "Application mortar". Since 801 is stored in 14 and retrieved from a dictionary, any errors in the character recognition results are corrected correctly.If the word recognition result for the key item is undecidable, the subsequent character codes Since it is not possible to test, all the accompanying character codes are inverted to negative.

When the verification is completed and the character code string shown in Figure 8 is rewritten (if there are no errors, there will be no change as a result).
, 20 output the same character code strings 801 and 802 onto the output line 50.

The above process is executed for each line of the form.

[Summary] As explained above, the apparatus of the present invention correctly corrects misrecognitions in character recognition results and outputs them.

The misrecognition rate can be reduced.

The results of this character recognition device are displayed, for example,
The final judgment can be made manually.

In other words, if a positive character code is output, it will be displayed normally, but if a negative character code is output, there is a high possibility of misrecognition, so you may want to change the brightness or color of the display, or change the gutter display. You can display special symbols next to the characters.
It can be corrected manually.

The features of the device of the present invention are that it does not require major changes as it can be installed after the conventional character recognition device, and that the verification processing section can be easily removed and the recognition section can operate as a conventional recognition device as is. Therefore, this verification processing section can be handled as an option.

Another feature of this method is that the measure of the closeness between any two characters, which is required in the method of searching a dictionary from a character code sequence containing errors during word recognition, is obtained from the similarity between standard patterns. . Therefore, a large amount of storage for storing proximity measures is unnecessary. In the embodiment described in this specification, the similarity calculation circuit 3o is provided in 10 in FIG. 5, but since 4 originally has the similarity calculation function, 4 may be slightly modified. Therefore, 30 can be included in 4, making it more efficient as a whole.

[Brief explanation of drawings]

FIG. 1 is a flowchart for explaining the invention in detail. FIG. 2 is a diagram of the structure of a word dictionary, and FIG. 3 is a diagram of character code strings corresponding to key items. FIG. 4 is a flowchart for explaining the word recognition (key item recognition) algorithm, FIG. 5 is a flag table, and FIG. 6 is a bit number conversion table. FIG. 7 is a block diagram of one embodiment of the present invention. Figure 8 is a diagram of the character code string resulting from character recognition.

Claims (1)

[Claims] 1. An input means for inputting an unknown character pattern, an output means for comparing the input unknown character pattern with a standard pattern and outputting the recognition result for each word, a storage means for storing word information, and the above output. a collation means for comparing and collating the word information with the recognition result for each word outputted by the means; and means for specifying the unknown character pattern for each word based on the collation result; . A character recognition device comprising means for distinguishing and displaying characters that are likely to be misrecognized from other characters based on the matching results.