JPH044632B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to a recognition method that can recognize written characters input in an online character recognition device used as an input device of an information processing device regardless of the stroke order.

(Background technology) Conventionally, in online character recognition methods, pattern matching is performed between unknown input characters and standard patterns using information on the stroke order and number of strokes of input characters that have been written, and the distance calculation results are calculated. Based on the recognition result, the one with the minimum distance is taken as the recognition result, and features indicating the stroke shape of unknown input characters (hereinafter referred to as stroke features), features indicating the positional relationship between strokes (hereinafter referred to as inter-stroke features), etc. are extracted. However, recognition methods have been adopted in which the similarity is calculated between these features and feature information of standard patterns stored in a character feature dictionary, and the one with the maximum similarity is used as the recognition result.

For example, taking the number "4" as an example, first, the two-stroke standard pattern group in the character feature dictionary is determined as the standard pattern group to be matched with the input character using the stroke count information of the input character. Next, the shape of the first input stroke "L" and the first stroke of the standard pattern is compared using the stroke order information of the input character, and the shape of "1" of the second input stroke is compared with the second stroke of the standard pattern. Further, the positional relationship between the first and second strokes is verified, and the sum of these verification results is taken as the degree of similarity between the input character "4" and the standard pattern. The above operation is repeated for all standard patterns of two strokes, and the one with the maximum similarity between characters is detected and used as a recognized character. Therefore, if the input character "4" is written in the stroke order of "/" and "L", the comparison with the standard pattern cannot be performed correctly, resulting in erroneous recognition. To solve this problem, the first stroke is "/" and the second stroke is "L" as a standard pattern that is opposite to the standard stroke order in the character feature dictionary.
4, or by exchanging the stroke order information between the strokes of the standard pattern and the strokes of the input characters, find the similarity for each and every combination of strokes, and calculate the similarity for each stroke of the standard pattern. By matching the strokes of the largest input character, a recognition method that does not depend on stroke order is adopted.

However, with these methods, as the number of images increases, the number of standard patterns increases, and matching with standard patterns requires a large amount of time.To avoid these problems, the device configuration must be complicated. It had some drawbacks.

(Problem of the Invention) In order to eliminate the above-mentioned drawbacks, the present invention provides a system that has a simple device configuration and can recognize input characters regardless of stroke order. First, after extracting the strokes of the input characters that have been written, the recognized characters are roughly classified using character information unrelated to the order of strokes, such as the number of strokes, and then the strokes of the input characters are compared with the stroke numbers of the standard pattern The stroke number of the input character is determined by using a stroke number dictionary that has stroke number determination information for each character whose stroke number can be determined, and a stroke number matching unit that matches the stroke information of the input character with the stroke number dictionary information. a character feature dictionary in which character feature information of a standard pattern to be matched is stored, in which character feature information of the stroke and inter-stroke feature information are extracted as character feature information of the unknown input character; By having a character feature matching unit that matches input character feature information with standard pattern character feature dictionary information, even if the same character has different stroke orders, it can be used to identify characters with multiple patterns of information in the character feature dictionary. The present invention is designed to allow recognition processing to be performed without any need for recognition, to recognize input characters regardless of stroke order, to simplify the device configuration, and to greatly reduce recognition processing time, which will be described in detail below.

(Explanation of the structure, operation, and effect of the first embodiment) FIG. 1 is a block diagram showing an embodiment of the present invention, in which 1 is a tablet, and 2 is a block diagram of an input character based on coordinate data from the tablet 1. stroke extraction,
and a stroke data extraction unit that extracts features of stroke data, and 3 a major classification discrimination unit that roughly categorizes recognized characters using character information such as the number of strokes that does not depend on stroke order and determines a standard pattern character group to be matched. , 4 is a stroke number dictionary in which stroke number determination information is stored for each character in order to assign a stroke number corresponding to the stroke number of the standard pattern to each stroke of an input character, and 5 is a stroke number dictionary of the stroke number dictionary 4. A stroke number matching unit for comparing dictionary information with the stroke of an input character, 6 a character feature dictionary storing character feature information of each character, 7 a comparison between the character feature information of the character feature dictionary 6 and the input character. This is a character feature matching unit that compares and matches feature information.

In the embodiment with the above configuration, the coordinate data of a handwritten input character inputted from the tablet 1 is sent to the stroke data extraction unit 2 after being subjected to known preprocessing such as noise removal, smoothing, and normalization. known stroke extraction based on the above-mentioned preprocessed character information in the order in which it was input.
Furthermore, feature extraction of the stroke is performed, and this input character information is output to the major classification discriminating section 3 and the stroke number collating section 5.

Based on the stroke data extracted by the stroke data extraction unit 2, the major classification discrimination unit 3 determines a standard pattern character group (candidate character group) to be matched, using simple information such as the number of strokes that is not related to stroke order. The recognized characters are roughly classified and the candidate character groups are sequentially output to the stroke number dictionary 4 and the character feature dictionary 6.
outputs the stroke number determination information of the candidate character to the stroke number matching unit 5 according to the input candidate character information. In the stroke number verification section 5,
The stroke information of the unknown input character extracted by the stroke data extraction unit 2 is compared with the stroke number determination information of the candidate character output from the stroke number dictionary 4, and each stroke of the input character and each standard pattern of the candidate character are compared. The stroke number of the input character is determined and output to the character feature matching section 7 so that the strokes of the input character are associated with each other, regardless of the stroke order of the input character. In accordance with the candidate characters output from the major classification discriminator 3, the character feature matching unit 7 compares candidate character features consisting of stroke features and inter-stroke features of the standard pattern of candidate characters output from the character feature dictionary 6, and strokes. Number verification section 5
The stroke characteristics of the input character and the inter-stroke characteristics are compared with the unknown input character characteristics by a method such as a known stroke analysis method based on the stroke number of the input character determined by the method, and the degree of similarity is calculated. This calculation result is stored in a candidate character similarity register (not shown).

In this way, all the candidate characters sequentially output from the major classification discriminator 3 are compared, and the character with the maximum similarity between the unknown input character written and the standard pattern of the candidate character group is determined as the recognition result.

The online character recognition method according to the present invention will be described below as an example where the three-stroke character "A" is written in the stroke order of "\", "/", and "-" as shown in Figure 2. This will be explained in more detail using an example of the stroke number dictionary shown in FIG.

The input character coordinate data from the tablet 1 is sent to the stroke data extraction section 2, where it is subjected to the above-mentioned known preprocessing, and then stroke extraction and stroke feature extraction are performed. The first input stroke is "〓", then the second input stroke is "/", and the third input stroke is "-".

The major classification discrimination unit 3 uses information such as the number of strokes to determine candidate characters of the standard pattern to be matched.
"G", . . . , etc. are sequentially output to the stroke number dictionary 4 and the character feature dictionary section 6 as candidate characters to be compared. The stroke number dictionary 4 uses the start point coordinates (Xns, Yns), end point coordinates (Xn _T , Yn _T ), etc. of the n-th input stroke of the input character as shown in FIG. 2, as shown in an example in FIG. 3. By using simple methods such as determining the size of the coordinate values, correspondence can be established by comparing the input stroke of the input character with the stroke of the standard pattern of the candidate character to be matched.
Stroke number determination information is stored for each candidate character. Further, the stroke number of the standard pattern is determined so that the stroke number of the standard pattern of each candidate character can be determined uniformly, as shown in FIG. 3, regardless of the normal order of strokes. When “A” is input as a candidate character to be matched into the stroke number dictionary 4, the first, second, and third input strokes of the input character shown in FIG. 2, “〓”, “/”,
“-” is the “A” in the stroke number dictionary 4, an example of which is shown in FIG.
The stroke number information is compared with the stroke number information.

As a result, the standard pattern for candidate character “A” is the first
The stroke is “-”, the second stroke is “〓”, and the third stroke is “/”, and the first stroke information of this standard pattern is “3 strokes
Since "Yn _T is the largest stroke", "-" corresponds to the stroke of the input character "A". Therefore, the stroke number 1 is assigned to the stroke "-" (third input stroke) of the input character "A". Similarly, the second stroke information of the standard pattern is the one with larger Xn _T among the remaining strokes, so the stroke number 2 is assigned to the stroke “〓” (first input stroke) of the input character “A”, and The remaining stroke is the stroke “/” of the input character “A” (second input stroke)
is given a stroke number 3.

The input character features made up of the stroke feature information of the n-th stroke and the inter-stroke feature information determined in this way are output to the character feature matching section 7. Further, the character feature dictionary 6 outputs character feature information of the standard pattern "A" to the character feature matching section 7 based on the candidate character "A" outputted from the major classification discriminating section 3. The character feature matching section 7 matches the input character feature information with the character feature information of the candidate character "A" to calculate the degree of similarity. As a result, the degree of similarity becomes very large. Regarding the next candidate character "E", the stroke number determination information of the candidate character "E" in the stroke number dictionary 4, an example of which is shown in FIG. -" is the stroke where Yn _T is the largest among the three strokes, the second stroke "-" is the one with the largest Yn _T among the remaining two strokes, and the third stroke "L" is the remaining stroke, so the input character " Each stroke of "A" is "-" (third input stroke), "/" (second input stroke), and "〓" (first input stroke).
Stroke numbers 1, 2, and 3 are given in the order of input strokes), and the strokes "-", "/", and "〓" of input character "A" are the same as "-" and "-" of standard pattern "E". ”, “L”, the stroke number (nth stroke) is determined and compared, and the degree of similarity between the input character and the candidate character “E” is calculated.
As a result, there is a clear difference between the second and third strokes, so the degree of similarity is small. Repeat the above operation for all candidate characters after "F",
The degree of similarity of each candidate character is calculated, and as a result, "A" is recognized as the character with the greatest degree of similarity.

As explained in detail above, since the strokes of unknown input characters and the strokes of the standard pattern of candidate characters can be easily matched, recognition processing can be performed regardless of the stroke order of the input characters, and the stroke order when writing the same character can be easily established. The present invention has the advantage that even if the characters are different, the character characteristic dictionary does not have a plurality of standard patterns of the same character, and input characters can be recognized regardless of the stroke order in which they are written. Further, according to this method, the stroke number dictionary 4 and the stroke number matching section 5
By having , in order to match the correspondence between input strokes and standard pattern strokes, as has been done in the past, stroke matching can be performed by considering the number of possible permutations of all mutual strokes. In addition, it is no longer necessary to apply conditions to some strokes and change their combinations to perform matching, and it is very easy to match the strokes of input characters with the strokes of the standard pattern of candidate characters. This has the advantage that the recognition processing time is extremely short and the device configuration can be extremely simplified.

In addition, in the above embodiment, when ranking strokes, the start point coordinates and end point coordinates are taken into consideration, but the center point coordinates, the rightmost (leftmost) coordinates, the uppermost (downward) coordinates, etc. are also included. can be set. In carrying out the present invention, it is important that input characters and candidate characters are ranked according to the same calculation rule, so that corresponding strokes of input characters and candidate characters can be uniquely distinguished from each other. It is guaranteed that the comparison can be made with a high degree of similarity. This is also important when comparing and matching the positional relationship between strokes, and it is determined based on the above order which two strokes of the candidate character correspond to the two strokes of the target input character. can be determined.

(Effects of the Invention) As is clear from the above description, the present invention can recognize input characters without being constrained by the stroke order of input characters in online character recognition where stroke order information of input characters is used as important information. can recognize,
Furthermore, it provides an effective means for shortening the recognition processing time, reducing the size of character feature dictionary information, and simplifying the device configuration.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a diagram showing an example of the input stroke order of a three-stroke input character, and FIG. 3 is a diagram showing an example of the contents of a 4-stroke number dictionary. 1... Tablet, 2... Stroke data extraction section, 3... Major classification discrimination section, 4... Stroke number dictionary, 5... Stroke number matching section, 6... Character feature dictionary, 7... Character feature matching section .

Claims (1)

[Claims] 1 In an online character recognition method that performs character verification by comparing and matching the character pattern of an input character written on a tablet with the character pattern of a prepared character stored in advance in a dictionary section based on the stroke shape and the position between the strokes. , a ranking is given to the strokes that make up a character, and this ranking and arithmetic rules for deriving the ranking are stored in the dictionary section in association with each character, and when the character is written on the tablet, the input Discrimination information on the number of strokes that does not depend on the stroke order is extracted from the characters, and based on this discrimination information, a group of characters close to the input character are extracted from the dictionary section to form a group of candidate characters, and the calculation rules for each candidate character are used to An online character recognition method characterized by assigning a rank to each stroke of the input character, and setting strokes of the same rank between the input character and a candidate character as strokes to be compared.