JPH0474756B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention easily and effectively recognizes, for example, a handwritten word consisting of n characters, and the attributes of a database, etc. indicated by the recognition result. This invention relates to a character recognition device that can obtain data.

[Technical background of the invention and its problems]

When recognizing a character, it is basically done by detecting the characteristics of that character, but when recognizing a word consisting of n characters, it is not enough to recognize each individual character and combine the results. There are many things.
Furthermore, if the recognition of a certain character is insufficient, it may be difficult to recognize the word. Therefore, in the past, a recognition device was configured as shown in Figure 1, and by taking advantage of the limited number of meaningful words, it was devised to recognize the word even when individual character recognition was insufficient. is applied. That is, a word consisting of a character string of n characters given through the character input section 1 is led to the recognition section 2 for character recognition. If this recognition is difficult, a plurality of candidate categories for the word are determined and provided to the word matching unit 3.
The word matching section 3 searches for combinations that are significant as words in the candidate categories from among the words registered in the word dictionary 4, and thereby selects and recognizes the correct word. According to such recognition processing, even if the recognition results for individual characters are incomplete, significant words can be selected by combining candidate categories, and as a result, effective word recognition is possible. Become. However,
When there are many candidate categories obtained through character recognition, the number of combinations becomes enormous, and there is a problem in that it takes an extremely long time to match words.

In addition, if you want to be able to obtain attribute information attached to recognized words in a character recognition device, what has been considered so far is
A dictionary in which words to be recognized are registered is prepared as a dictionary used for comparison with candidate categories, and a table in which attribute data corresponding to the words is stored is separately prepared.

In other words, conventionally, for example, if we take a person's name as a recognition target word and an address as attribute data, a database that stores the person's name/address in correspondence already exists, so a character recognition device is constructed separately from this database. One idea has been to search the existing database using words (personal names) obtained after performing character recognition using a word dictionary that stores only personal names.

However, the methods that have been considered so far have the disadvantage that the database to be managed consists of two databases: a word dictionary used for character recognition and a table used for searching attribute data, which makes management cumbersome. It is necessary to go through the complicated steps of searching to identify the recognition target word in the word dictionary, specifying the table to read the attribute table after identifying it, and searching the table.
The disadvantage is that it takes time to do it once.

[Purpose of the invention]

The present invention was made in consideration of these circumstances, and its purpose is to perform effective word recognition by simply and quickly comparing dictionary words and candidate categories, and at the same time, to perform word recognition effectively. It is an object of the present invention to provide a character recognition device which can also effectively obtain attribute data of words.

[Summary of the invention]

The present invention provides a means for recognizing each character of an input word given as a character string to obtain a candidate category for each character, and a method in which a recognition target word and attribute data corresponding to the recognition target word are arranged in advance as adjacent pairs. A word dictionary in which a plurality of sets are registered, a means for searching only the recognition target word excluding the attribute data from the word dictionary, and comparing and matching the searched recognition target word with the candidate category, and recognition in which the comparison and verification have been performed. Means for calculating the degree of matching of the target word with the input word from the candidate ranking of the candidate category or the degree of similarity to the characters of the input word and selecting a recognition target word with a high degree of matching as a candidate word; It is constituted by means for reading attribute data corresponding to the recognition target word from the word dictionary.

〔Effect of the invention〕

Therefore, according to the present invention, recognition target words and corresponding attribute data are paired in a word dictionary used for character recognition, and these are registered so that they are adjacent to each other, and each recognition target word and candidate category are compared and matched. edge,
By adopting a new configuration in which attribute data is skipped and only the words to be recognized are searched, the number of databases to be managed can be reduced to one, simplifying management control and identifying the words to be recognized in the word dictionary. Since it is only necessary to read out the attribute data registered adjacent to this, the search can be performed only once, and the time can be significantly reduced.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 2 is a schematic configuration diagram of the embodiment device. The input word given as a string of n characters is
Each character is recognized by the recognition unit 11,
For each, r candidate categories (character codes) are determined. These candidate categories are determined for each word character position from the first place to the rth place according to the candidate ranking. At this time, it is also possible to calculate the degree of similarity for characters in each candidate category, that is, the degree to which the standard character pattern of the candidate category and the input character pattern are similar, and to treat this degree of similarity and the candidate category as one. good. Thus, the number of recognition target words obtained by combining r candidate categories for each of the n characters found by the recognition unit 11 is r ⁿ . These candidate categories are then provided to the word matching unit 13 via the editing unit 12.

On the other hand, in the word dictionary 14, Ln words each consisting of n characters and attribute data attached to the words are registered in advance in correspondence with each other. These words are sequentially read out and provided to the word matching section 13, where they are subjected to recognition processing.

Now, the word matching section 13 is basically configured as follows. r candidate categories for n characters recognized through the recognition unit 11, that is, n
The ×r candidate categories are stored in the candidate character register 21. Further, words read out from the word dictionary 14 under the control of the address counter 22 are stored in the word dictionary register 23. Each character of the word stored in this word dictionary register 23 is supplied to a comparator 24, and under the control of a register counter 25, it is compared in accordance with the candidate category read out sequentially from the candidate character register 21 and the word character position. Ru. This comparator 24 matches word characters and candidate categories for each word character position. The matching detection information is then given to a matching degree calculating section 26, which calculates the matching degree of the word stored in the register 23 and used for comparison with the input word. This degree of matching calculation is performed, for example, by determining the candidate ranking of the matched candidate category for each word character position, and calculating the total value of the candidate ranking information. Furthermore, when the degree of similarity has been determined for candidate categories as described above, the degree of coincidence may be determined as the sum of the degrees of similarity of candidate categories whose matches are detected for each word character position. Note that if a matching candidate category is not found at any word character position through this matching comparison, it may be treated as if there is no matching for that word. Then, such word matching calculation is repeated every time a word is read from the word dictionary 14, and the matching degree is determined for all Ln words. The information on the degree of matching of each word with respect to the input word thus obtained is led to the sorting processing section, and is sorted in descending order of the degree of matching. Then, x words with a high degree of matching are given to the editing unit 12 as candidate words. The editing unit 12 comprehensively edits these candidate words and the recognition information of the input word, determines them, obtains a recognition result, and obtains the word and attribute data attached to the word.

By the way, such a verification process is specifically performed as follows. FIG. 3 shows an example, in which the candidate character register 21 is constituted by a shift register having a storage area of n rows and r columns. Thus, the candidate categories recognized for each of the n characters are stored in sequence over r columns in accordance with the candidate rankings, with the word character positions corresponding to the n rows. These stored candidate categories are then read out in parallel column by column under the control of the counter 25 and provided to the comparator 24.
Further, the comparator 24 is provided with character data of each word from the register 23 in parallel in correspondence with word character positions. This allows simultaneous comparison of candidate categories and word characters at each word character position. Then, information from the word character position for which a matching result is obtained is given to the matching degree calculation section 26, and the calculated value of the counter 25 at that time is taken in as candidate ranking information. When one comparison is completed, the counter 25 is incremented and the next column of candidate categories is given to the comparator 24.
Match detection is performed in the same manner. This matching detection is repeated r times, thereby completing the matching process for the words stored in the register 23.

At this time, the match calculation unit 26 has obtained a count value indicating the candidate category in which a match has been detected for each word character position, that is, information on the candidate ranking, and according to this information, The degree of matching of a word to an input word is determined. The sorting processing unit 27 recognizes the word for which the degree of matching has been determined from the count value of the address counter 22 at this time, and selects the word with the high degree of matching.
It is stored as a match value and a count value. Then, the matching degree of the next input word is compared with the matching degree of the previously input word, and the word information with a low matching degree is discarded, and the sorting process is performed by rearranging the word information. It's on.

In this case, the similarity information corresponding to the candidate category stored in the register 21 is stored in a separate storage area, and the matching degree calculation is performed according to the similarity information instead of the above-mentioned candidate ranking. You can do it like this.

Further, in the configuration shown in FIG. 3, the collation process is executed simultaneously in parallel for each word character position, but it may be executed serially as shown in FIG. 4. That is, candidate character register 2
1 has n×r storage areas, and the candidate categories are sequentially stored in these storage areas. Here, an example is shown in which similarity information is also stored together with candidate categories. Then, the candidate categories are sequentially read out under address control of the counter 25 and supplied to the comparator 24, and each character of the word stored in the register 23 is sequentially read out via the selector 28. In this case, the counter selects the first character and sequentially reads candidate characters at that character position to detect a match, and then selects the second character and performs match detection for the candidate category at that character position. 25 address control and selector 2
8 selection controls may be synchronized. Even in this case, the matching process can be performed in the same way as in the previous example.

FIG. 5 is a diagram showing the concept of the above-mentioned recognition process, and here shows an example in which a person's name word "Kurosawa" is given as an input word. Therefore, in the word dictionary 14, along with the word 14a to be recognized,
Attribute data 14b attached to the word 14a is registered in advance. Among these words, the word 14a is subjected to the recognition process for the input word. When this word 14a is obtained as a recognition result as described above, this word 14a
At the same time, the attribute data 14b is read out. This reading process is performed by the editing section 12.

Note that the attribute data 14b is, for example, the word 14.
When a is given as a person's name word, it is given as data such as the address, telephone number, age, etc. This attribute data 14b is used to search and manage databases such as member lists and customer ledgers. Further, when the word 14a is given in kana characters, kana/kanji conversion can be performed by giving the attribute data 14b as the corresponding kanji.

As explained above, according to the present invention, recognition target words and corresponding attribute data are grouped in a word dictionary used for character recognition, are registered so that they are adjacent to each other, and each recognition target word is compared with candidate categories. By adopting a new configuration in which attribute data is skipped and only words to be recognized are searched during verification, the system can manage only one database. This simplifies management control, and when a word to be recognized in the word dictionary is identified, the attribute data registered adjacent to it can be read out, so the search only needs to be done once, which significantly saves time. It will be possible to shorten it.

Note that the present invention is not limited to the above embodiments. For example, word classification information may be provided as the attribute data. It is also possible to provide information on the frequency of occurrence of the word as attribute data.
In short, the present invention can be implemented with various modifications without departing from the gist thereof.

[Brief explanation of drawings]

FIG. 1 is a schematic configuration diagram showing an example of a conventional device, FIG. 2 is a schematic configuration diagram of an embodiment of the device of the present invention, and FIG.
4 and 4 are diagrams each showing a basic configuration example of the word matching section, and FIG. 5 is a diagram showing the concept of recognition processing. 11... Recognition unit, 12... Editing unit, 13... Word matching unit, 14... Word dictionary, 21... Candidate character register,
22... Address counter, 23... Word dictionary register, 24... Comparator, 25... Register counter, 26... Matching degree calculation section, 27... Sorting processing section,
28...Selector, 14a...Word, 14b...Attribute data.

Claims (1)

[Scope of Claims] 1. Means for recognizing each character of an input word given as a character string to obtain each candidate category; a word dictionary in which a plurality of sets of words are registered; a means for searching only the recognition target word excluding the attribute data from the word dictionary; and comparing and matching the searched recognition target word with the candidate category; means for calculating the matching degree of the matched recognition target word with the input word from the candidate ranking of the candidate category or the similarity to the characters of the input word, and selecting a recognition target word with a high matching degree as a candidate word; A character recognition device comprising means for reading out a selected recognition target word and attribute data corresponding to the recognition target word from the word dictionary.