JPH0475556B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention mainly involves the configuration of a recognition dictionary for reading printed characters in a plurality of fonts, and the use of the recognition dictionary to determine the category and font to which an input character belongs. This invention relates to a character reading device.

[Prior art]

Conventionally, the recognition methods of character reading devices for printed characters include a pattern matching method in which the character patterns themselves are superimposed and the one that best overlaps is selected as a candidate, and a method in which the line segments that make up the character are determined from the character pattern. There is a feature matching method that extracts features and selects those with the most similar features as candidates. In character reading devices that use these recognition methods, dictionary configurations for reading multiple fonts include one that prepares an average of all font shapes, and one that prepares a dictionary for each font for each category. . In the former case, the recognition dictionary is composed of average features, which makes it impossible to perform highly accurate recognition and also makes it impossible to recognize fonts. The latter has the disadvantage that the memory amount of the recognition dictionary increases and the identification process becomes slow. Further, none of the conventional character reading devices has a function of outputting the font of the character to be read together with the character code.

[Summary of the invention]

In order to solve these drawbacks, this invention
We focused on the fact that many of the same features that appear in character patterns exist even if the fonts (Mincho font, Gothic font, etc.) are different as long as they are in the same category (what the character is called). By classifying and registering characters into common features, which are the features that commonly appear in each font, and individual features, which have features that appear uniquely in each font, the size of the recognition dictionary can be saved, and the size of the recognition dictionary can be reduced. It is designed to recognize fonts at the same time. Hereinafter, this invention will be explained with reference to the drawings.

[Embodiments of the invention]

FIG. 1 shows an example of the configuration of a character reading device according to the present invention. In this figure, numeral 1 is a form on which input characters to be read are written. 2 is a preprocessing unit that cuts out characters written on form 1 one by one. Reference numeral 3 designates a feature extraction section, 4 a recognition section, 5 a recognition dictionary section in which features are stored in advance, 6 a control section, and 7 an output terminal.

Next, the operation will be explained. After the characters written on the form 1 set in the device are photoelectrically converted line by line, the preprocessing unit 2 extracts the characters one by one according to the signals from the control unit 6, and the feature extraction unit 3 extracts the characteristics of each character. are extracted and these feature data are sent to the identification unit 4.
sent to. In the identification section 4, for example, in a similarity calculation method using a linear discriminant function, the features of the input character are compared with the features prepared in advance in the recognition dictionary section 5,
The degree of similarity is calculated, and the category showing the highest degree of similarity or the category having the degree of similarity above a certain threshold value is output.

Next, the method of configuring the recognition dictionary section 5 and the method of recognizing categories and fonts in the identification section 4 will be explained in detail with reference to FIG. In FIG. 2, a is a conceptual diagram of a conventional recognition dictionary configuration, and b is a conceptual diagram of a recognition dictionary configuration according to the present invention. Here, for the sake of simplicity, a case where two fonts are to be read will be described.

In a, for the same category C ₁ , the constituent elements of feature F ₁ of font #1 are (f ₁ , f ₂ , f ₄ , f ₆ ,...
...f _o-2 , f _o ), and the constituent elements of feature F ₂ of font #2 are (f ₂ , f ₃ , f ₅ , f ₆ , ... f _o-2 , f _o-1 ) ,
If they are combined as is, the size of the recognition dictionary section 5 will be approximately doubled.

Therefore, as shown in figure b, font #1 and font #
F ₉ (f ₂ , f ₆ ,
...f _o-2 ), the features (individual features) of font #1 only
F' ₁ = (f ₁ , f ₄ , ... f _o ), classifying the features (individual features) of font #2 only as F' ₂ = ( f ₃ , f ₅ , ... f _o-1 ), They are registered separately in the recognition dictionary section 5. Do this for all categories.

When performing identification using the recognition dictionary unit 5, the characteristics of the input character and the characteristics of the recognition dictionary unit 5 (F ₀ +F′ ₁ ), (F ₀ +
F′ ₂ ), and the font with the highest degree of similarity is selected as a candidate for the font type, thereby recognizing the font and simultaneously outputting category and font information. At this time, it is only necessary to check the input character features and the common feature F ₀ once, and as shown in Figure 2 a, F ₁ , F ₂
Identification processing can be performed faster than when matching is performed with all the features of .

It goes without saying that fonts can be recognized with even higher accuracy if a dictionary for font recognition that uses features such as line width ratio and the presence or absence of decoration is used in combination with the recognition dictionary unit according to the present invention. .

〔Effect of the invention〕

As explained above, the present invention configures a recognition dictionary section by registering common features and individual features separately, and compares these features with features extracted from input characters to determine which font in which category. Since the category and font information is obtained by determining whether the individual characteristics of There are advantages that can be applied.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG. 2 is a conceptual diagram illustrating the configuration of a recognition dictionary section according to the present invention. In the figure, 1 is a form, 2 is a preprocessing unit, 3 is a feature extraction unit, 4 is an identification unit, 5 is a recognition dictionary unit, 6 is a control unit,
7 is an output terminal, a is a conventional recognition dictionary configuration, b is a recognition dictionary configuration according to the present invention, F ₁ , F ₂ , F ₀ ,
F′ ₁ and F′ ₂ are each in the same category C ₁ ,
F ₁ is a feature of font #1, F ₂ is a feature of font #2, F ₀
is a common feature of font #1 and #2, _F'1 is a feature of font #1 only, and _F'2 is a feature of font #2.

Claims (1)

[Claims] 1. In a character reading device that reads input characters containing a mixture of multiple types of fonts, it is possible to distinguish between common features that appear in the same category even if the fonts are different, and individual features that appear uniquely for each font in that category. A recognition dictionary section that is registered separately, a feature extraction section that extracts each feature from the input character, and each feature in the recognition dictionary section and each feature extracted from the input character of the feature extraction section. A character characterized by comprising an identification unit that determines whether the common feature matches, and then determines which individual feature of the font in that category matches the character, and outputs the category and font information when comparing. reading device.