JPS6292088A - Recognizing device for on-line handwritten character - Google Patents

Abstract

PURPOSE:To recognize accurately the handwritten characters by registering previously the information on the degree of resemblance in response to the basic stroke and then using this registered information for the recognition of handwritten characters. CONSTITUTION:When the handwritten characters are supplied from a handwritten character input means 1, a quantization processing means 2 calculates the handwritten strokes of the handwritten characters and informs the result of calculation to a character resemblance degree calculating means 3. The means 3 retrieves a basic stroke resemblance degree dictionary 4 for the calculation of the degree of resemblance and also retrieves the characters corresponding to the result of calculation. By receiving the information on the retrieving result of characters, an output character candidate selecting means 5 delivers the minimum degree of resemblance, for example, as the handwritten characters. In such a way, the handwritten characters can be accurately recognized.

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] In an online handwritten character recognition device that recognizes characters input by hand, quantization processing means calculates handwritten strokes of characters input by hand.

At least a basic stroke approximation dictionary in which approximation information is registered in association with basic strokes.

The approximation is used to recognize handwritten characters.

[Industrial application field]

The present invention relates to an online handwritten character recognition device configured to recognize characters input by handwriting by searching for similarity information registered in advance in association with basic strokes. .

[Problems to be solved by conventional technology and invention] Conventional 9
The device that recognizes handwritten characters online is 1.
The handwritten strokes of the handwritten characters were compared with the basic strokes to be compared, and the degree of similarity was calculated to recognize the handwritten characters. Therefore, if some of the handwritten strokes input by handwriting are slightly different from the handwriting strokes when the correct characters are input by handwriting due to the habits of the person inputting the handwritten characters, the handwriting strokes input by handwriting will no longer be recognized correctly. The problem was that I couldn't get it.

[Means for solving problems]

In order to solve the above-mentioned problems, the present invention registers similarity information in advance in association with basic strokes.7 When recognizing handwritten characters, the registered similarity information is used to recognize handwritten characters. By adopting this configuration, even handwritten characters with peculiarities can be recognized.

FIG. 1 shows the basic configuration of the present invention. In the figure, 1 is a handwritten character input means, 2 is a quantization processing means, 3 is a character similarity calculation means, 4 is a basic stroke similarity dictionary, and 5 is an output character candidate selection means.

In FIG. 1, 3 handwritten character input means 1 are for inputting 1 handwritten character, and 2 are, for example, a tablet.

The quantization processing means 2 calculates handwritten strokes of input handwritten characters.

The character similarity calculation means 3 searches the basic stroke similarity dictionary 4, reads basic strokes with a predetermined degree of approximation, and calculates the degree of similarity.

The basic stroke approximation dictionary 4 is designed to register basic strokes that are more similar to the basic stroke. It is designed to be registered in the form of things, etc.

The output character candidate selection means 5 selects characters to be output based on the calculated degree of similarity.

[Effect]

When a handwritten character is input from the three handwritten character input means 1 using the configuration shown in FIG.
is calculated and notified to the nine character similarity calculation means 3. Upon receiving this notification, the character similarity calculation means 3 searches the basic stroke similarity dictionary 4 to calculate the similarity, and searches for a character corresponding to the calculated result. The output character candidate selection means 5 that received the notification of the searched character result 2
For example, the one with the smallest value of near A or higher is recognized as a handwritten character and output.

As explained above, even if the handwritten stroke of a part of the character input by handwriting does not match the correct basic stroke, the habits of the nine people are investigated in advance and the basic stroke corresponding to this habit is used as similarity information in advance. By registering, it becomes possible to accurately recognize handwritten characters.

〔Example〕

FIG. 2 shows the configuration of one embodiment of the present invention. In the figure.

1-1 is a tablet; 3-1 is a character similarity calculation means;
Reference numeral 6 represents a character coordinate reading means, 7 a normalization processing means, 8 a basic stroke extraction means, 9 a basic stroke dictionary, and 10 a character dictionary. Note that 2°4.5 in Figure 2 has the same effect as that shown in Figure 1.

In FIG. 2, a tablet 1-1 is used to input handwritten characters using a pen.

The character coordinate reading means 6 reads the coordinate values (x, y) of the trajectory of the pen input by handwriting from the tablet 1-1.

2).

The normalization processing means 7 normalizes the input handwritten characters to a predetermined size by reducing or enlarging them.

The basic stroke extracting means 8 searches the basic stroke dictionary 9 for the handwritten stroke of one handwritten input character and calculates the degree of similarity A or higher.

The basic stroke dictionary 9 is a dictionary in which correct basic strokes of characters are registered.

The character similarity calculation means 3-1 calculates a basic stroke corresponding to one handwritten stroke, as described above.

The basic stroke similarity dictionary 4 is read out, the degree of approximation is calculated, and the character corresponding to the combination of basic strokes adopted for the calculated degree of approximation is searched and read out from the 5-character dictionary 10.

The character dictionary 10 is a register of characters corresponding to combinations of basic strokes.

Output character candidate selection means 5 selects a predetermined character as a handwritten character from the recognized character candidates notified.

This is what is output.

As explained above, according to this embodiment, handwritten characters are recognized by comparing the basic strokes that constitute a correct character using the basic stroke extraction means 8, and the recognition is performed using the basic strokes. If not,
Furthermore, by adopting a configuration in which handwritten characters are recognized by comparing basic strokes corresponding to one person's habits.

We are trying to improve the recognition rate of handwritten characters.

Next, with reference to FIGS. 3 and 4, the operation when recognizing characters using similarity information will be described in detail. Third
The figure shows an example of a basic stroke approximation dictionary ξ7. FIG. 4 shows an example of identifying basic stroke approximations.

The stroke number shown in the upper part of FIG. 3 represents the address of one basic stroke.

The pattern shown at the top of the figure represents a basic stroke pattern.

The approximate stroke numbers shown at the top of the figure are approximation degree 0, approximation degree 1. and the degree of approximation is 2.

Each represents the rank of approximation to the pattern indicated using the stroke number. This degree of approximation 0°1.2 is 1
The handwritten characters are determined by studying the habits of people in advance, and the degree of approximation 0 is the most frequent, and the degree of approximation 1 is the most frequent.
is the next most common, and the degree of approximation 2 is the next most common.

Figure 4 shows an example of identifying the degree of approximation of basic strokes.
Figures (a) and (b) show the correct handwritten character "Do" and the basic strokes at that time (stroke numbers 1 10.l, 6 in Figure 3).

Figure 4 (C), (D), (E) and (F) are handwritten characters with quirks and approximate basic strokes (stroke numbers 1.9, 1.6. and handwritten characters with similar quirks). and stroke numbers 1.1°1.6) are shown, respectively.

FIG. 4(a) shows the correct character "door", and when this correct character "door" is input from the tablet 1-1, the quantization processing means 2 converts it into a handwritten character as shown in FIG. 4(b). Calculate the stroke. This handwritten stroke (stroke numbers 1, 10°, 6 in Figure 3) is exactly the same as the basic stroke combination (stroke numbers 1, 10, 1.6), so it is easy to calculate. can recognize the character ``to''.

FIG. 4(C) shows a character with a quirk, and the part indicated by (a) in FIG. 9 is oblique, and this character with a quirk is inputted from the tablet 1-1.

The quantization processing means 2 calculates handwritten strokes as shown in FIG. 4(d). These handwritten strokes (stroke numbers 1, 9, 1.6 in Figure 3) are.

This is different from the correct basic stroke combination (stroke numbers 1, 10, 1.6) of the character "door", so it will not be recognized. For this reason, one character similarity calculation means 3
searches the illustrated basic stroke similarity dictionary 4 in FIG. 3, searches the basic stroke similarity dictionary 4 based on the basic strokes shown in FIG. Since the degree of approximation is O, the combination of basic strokes (rotoke number 1, 10.1.6) can also be extracted as a matching target and recognized as the character "door". do.

Similarly, Figure 4 (e) has a peculiar character.

A part (vertical part) of the character indicated using (in the middle of the figure) is missing. For this character, as shown in FIG.
Therefore, it is possible to extract stroke numbers (stroke numbers 1, 1, 1.6) as matching targets and recognize the two characters "to".

As explained above, a character with a peculiar part,
By investigating this information as similarity information and registering it in advance for basic strokes, it is possible to improve the recognition rate of characters input by hand.

〔Effect of the invention〕

As explained above, according to the present invention, similarity information is registered in advance in association with basic strokes, and when recognizing handwritten characters, the registered similarity information is used to determine the degree of similarity. Since it uses a configuration that calculates and recognizes handwritten characters, it is possible to recognize even handwritten characters with irregularities.

It is possible to improve the recognition rate.

[Brief explanation of drawings]

FIG. 1 shows the basic configuration of the present invention, FIG. 2 shows the configuration of one embodiment of the invention, FIG. 3 shows an example of a basic stroke approximation dictionary, and FIG. 4 shows an example of basic stroke approximation identification. In the figure, 1 is a handwritten character input means, 2 is a quantization processing means, and 3
4 represents a character similarity calculation means, 4 represents a basic stroke similarity dictionary, and 5 represents an output character candidate selection means. Patent Applicant Alps Electric Co., Ltd. Representative Patent Attorney Hiroshi Mori 1) (and 3 others) Book 71!
cron (c) (d)
(Sai, N) (He) Basic stroke approach If S block ~'￥su 4 multi゛j Fig. 4 $ 3 Fig.

Claims (1)

[Claims] An online handwritten character recognition device that recognizes characters input by hand, comprising: a quantization processing means for calculating handwritten strokes of characters input by hand; and quantization processing means for approximating the basic strokes in association with the basic strokes. At least a basic stroke similarity dictionary in which similarity information indicating strokes is registered, and a character similarity calculation means for searching this basic stroke similarity dictionary and extracting strokes that are similar to handwritten strokes, An online handwritten character recognition device characterized in that it is configured to recognize characters input by hand based on the results obtained by the character similarity calculation means.