JPH0337231B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (A) Field of Industrial Application The present invention relates to a character recognition system for recognizing handwritten Chinese characters, and more particularly to a method for determining the recognition ranking of candidate character type categories.

(b) Conventional technology Generally, in a character recognition system, the degree of similarity between a feature pattern extracted from an input character pattern and a standard feature pattern for each character type category registered in advance in a dictionary section is calculated, and Select n candidate character type categories. Then, the candidate character type category with the highest degree of similarity is output as a recognition result, and in order to correct misrecognition, the selected n candidate character type categories are ranked from 1st to nth in order of similarity. Decide on the recognition ranking up to the first rank.

However, as mentioned above, using only similarity to determine the recognition ranking often results in false recognition, so after selecting multiple candidate character categories based on similarity, some post-processing is performed to determine the recognition ranking. Now I can think of a way to do that.

Conventionally, as post-processing, JP-A-59-
As disclosed in Japanese Patent Publication No. 32082, it performs grammatical processing, and as in Japanese Patent Application Laid-open No. 59-27381, it determines whether the characters before and after the recognized character are kanji, katakana, or hiragana. Something was proposed.

(c) Problems to be solved by the invention In the conventional technology, grammatical processing is performed as post-processing, so the knowledge section such as a grammatical dictionary becomes enormous, and furthermore, the processing content becomes extremely complicated. Furthermore, with the method of determining whether the preceding and succeeding characters are kanji or katakana, if the selected candidate character category is only kanji or hiragana, it cannot be expected to improve the recognition rate. .

Therefore, in order to improve the recognition rate in a short processing time without requiring a huge knowledge department, the applicant set priorities for each character type category according to the learning stage or frequency of school education. I then tried to use this priority to determine the recognition ranking of multiple candidate character type categories obtained by similarity calculation, but since the recognition rate changes depending on the balance between similarity and priority, I was wondering how to balance this. The problem has become how to set it up properly.

(d) Means for solving the problem The present invention sets a priority for each character type category, and for candidate character type categories whose similarity is greater than a predetermined value, the recognition order is changed based on the priority. Instead, the recognition order is determined by changing the recognition order based on the priority only for candidate character type categories smaller than a predetermined value.

(E) Effect In the present invention, for those with a large degree of similarity,
The priority is ignored and the recognition ranking is determined only by the similarity, and for items with a small similarity, the ranking is determined by the priority. Therefore, candidates for feature patterns that are very similar to the extracted feature pattern Character type categories have the same recognition order as before, and for candidate character type categories that are not very similar, the priority is taken into account and the recognition rate improves.

(F) Embodiment Figure 1 is a block diagram of a character recognition system to which the present invention is applied, and 1 is a character observation system that reads characters written on an input manuscript and outputs the reading results as a binary character pattern. Part 2 is a feature extraction part that extracts a feature pattern from an input character pattern;
Reference numeral 4 indicates a dictionary section that stores standard feature patterns for each character type category, and 4 indicates a pattern matching section that matches the extracted feature patterns with the standard feature patterns and calculates the degree of similarity between the two patterns.

The character categories in the dictionary section 3 are divided into categories according to frequency or learning stage of school education, and each category set is given a priority. For example, as shown in Figure 2, category set 1 (3a) is the character category learned in the first to third years of elementary school, category set 2 (3b) is the character category learned in the fourth to sixth grade of elementary school, and category set 2 (3b) is the character category learned in the 4th to 6th years of elementary school. All character categories are divided into three category sets, such as category set 3 (3c), and priorities 0 to 2 are assigned to category sets 1 to 3 in order.

The pattern matching unit 4 includes three calculation units 4a to 4c corresponding to category sets 1 to 3, respectively, and each calculation unit selects n candidate character type categories from each category set in descending order of similarity. is selected, and its character type code and the degree of similarity as a calculation result are stored in the candidate memory 5. On this occasion,
The calculation section adds the priority of the corresponding category set to the character type code and similarity, and these three pieces of information are stored in the candidate memory 5 as information for each candidate character type category. In this way, the candidate memory 5 stores n candidate character type categories from each category set, for a total of 3n candidate character type categories.

Further, in FIG. 1, reference numeral 6 denotes a knowledge section that stores the relationship between similarity and priority, and 7 refers to the contents of this knowledge section 6 to determine the 3n candidate character type categories stored in the candidate memory 5. It is a clustering control processing unit that determines the recognition ranking of the top n characters among them and stores the character type codes in the result memory 8 in order of recognition ranking, and the answer output control unit 9 determines the character type with the first recognition ranking. The code is output as a recognition result to a character display device such as a word processor or a personal computer, and the type of character is displayed. Then, when the operator gives an instruction of misrecognition, the answer output control section 9 sequentially outputs the second and lower character type codes, and controls the output so that the correct recognition result is displayed.

Next, the processing of the clustering control processing section 7 and the contents of the knowledge section 6 will be explained in more detail.

In this embodiment, the city block distance d is used as the degree of similarity, and it is assumed that the smaller this distance is, the greater the degree of similarity is. Threshold values D〓, D〓, D〓 (D〓<D〓<D〓) are stored. As shown in the flowchart of FIG.
From among the 3n candidate character categories stored in , n candidate character categories are selected in descending order of distance.
M ₁ , M ₂ , . . . , M _o are selected and stored in the working memory 7a. Then, by comparing the distances d ₁ , d ₂ , ..., d _o of each of these n candidate character type categories with multiple thresholds D〓, D〓, D〓,
The candidate character types are classified into categories M ₁ to M _o . That is, as shown in FIG. 4, A, B, C, D
divided into four classes. As shown in FIG. 4, the knowledge unit 6 stores in advance whether or not the order of priority can be changed in each class.
In the and C classes, it is specified that the order of priority is to be changed only within the same class. Then, the clustering control processing unit 7
The recognition order is determined by referring to the contents of

Here, the character type code is M ₁ , the city block distance d ₁ , and the priority is P (I=A, B, C..., P=0,
1, 2) candidate character type categories (M ₁ , d ₁ , P)
Now, suppose that the top five candidate character type categories with the highest degree of similarity are selected from each of category sets 1 to 3, and each city block distance is The relationship of d ₁ is d _V
Assuming that <d _P <d _A <d _Q <d _S <..., the working memory 7a of the clustering control processing section 7 has the following information:
Five character type codes M _V in the order shown in Figure 5 (b)
~M _S is selectively stored.

Therefore, for example, if the city block distances d _V and d _P of character type codes M _V and M _P are sufficiently small d _P < D〓, and D〓 < d _A and d _B < D〓, clustering control The processing unit 7 classifies the character type codes M _V and _MP into A class, and the character type codes M _A , M _Q , and M _B into B class, and classifies the character type codes M _V and M B of A class.
For M _P , the recognition order is not changed based on priority, and B rank character type codes M _A , M _Q ,
Since the recognition order for M _B is changed based on the priority, as shown in Figure 5 (c), even if the priority is low such as "2" or "1", the character type codes M _A and M have a high similarity. _Q was determined to be the 1st and 2nd place,
In class B, where the similarity is small, character type codes M _A and M _B with high priority of “0” are at the top, and are ranked 3rd overall.
and 4th place.

In addition, there are no candidates with extremely small distances, and D〓<d _V ,
If d _A < D < d _Q and d _B < D, the clustering control processing unit 7 sets the character type codes M _V , _MP ,
M _A is classified into B class, character code M _Q and M _B are classified into C class, and the recognition order is swapped based on priority within each class, so the distance is The character type code M _A with the highest priority is determined to be the first, even if it is slightly smaller.
Character type codes M _V with low priority are ranked lower.

In this way, for character type categories with high similarity, the recognition order is determined only by similarity, ignoring priority, as before, and for character type categories with low similarity, priority is taken into account. A determination of the ranking will be made.

(G) Effects of the Invention According to the present invention, for candidate character type categories whose similarity is large enough that the recognition ranking based on the similarity calculation result is sufficiently reliable, the recognition ranking is exactly the same as before, and the recognition result is Regarding ambiguous candidate character type categories, the recognition rate is improved because the ranking is determined taking priority into consideration.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a character recognition system to which the present invention is applied, FIG. 2 is an explanatory diagram showing the contents of a category set, FIG. 3 is a flowchart showing part of the processing contents of the clustering control processing section, and FIG. The figure is an explanatory diagram showing the contents of the knowledge section, and FIG. 5 is an explanatory diagram showing a specific example of recognition ranking determination. Explanation of main figure numbers 3...Dictionary section, 4...Pattern matching section, 5...Candidate memory, 6...Knowledge section, 7...
Clustering control unit, 8... Result memory, 9... Answer output control unit.

Claims (1)

[Claims] 1. In a character recognition system that calculates the degree of similarity between a feature pattern extracted from an input character pattern and a standard feature pattern for each character type category registered in advance in a dictionary section, and selects a plurality of complementary character type categories, the above-mentioned A priority is set for each of the character type categories, and among the plurality of Hou complementary character type categories, for the Hou complementary character type categories whose similarity is greater than a predetermined value, the recognition order is not replaced according to the priority, A recognition order determining method, characterized in that the recognition order is changed based on the priority for the Hou-complementary character type category that is smaller than the predetermined value.