JP2882056B2 - How to identify specific patterns - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for identifying a handwritten specific pattern.

[0002]

2. Description of the Related Art Conventionally, as a method of identifying a pattern by a computer having a left parenthesis "and a right parenthesis" indicating a range designation in a plate making machine, a pattern / character
A matching method is known. By comparing a template, which is a standard pattern prepared in advance, with the pattern of the input data, and associating the input data pattern with the pattern class of the template that overlaps the most, it is possible to identify which of the two patterns it is. I was doing.

[0003]

However, in order to perform the pattern matching method for pattern feature extraction, it is necessary to prepare a standard pattern in advance, and a storage location for the standard pattern is required.

[0004] Also, because of handwriting, the size of characters,
Processing for inclination is also required. When the pattern matching method is used, a pattern matching process, a normalization process for the size of a character, and a character rotation process are required for the inclination of the character.

[0005] For these reasons, the program becomes complicated and the capacity becomes large, so that there is a problem that the processing speed is slow.

An object of the present invention is to provide a method for identifying a specific pattern which can perform the above-described feature extraction of the pattern by a simple process in a short time.

[0007]

In order to achieve this object, a specific pattern identifying method according to the present invention comprises a pixel memory unit for storing a pattern to be identified, and a pattern stored in the pixel memory unit. , A raster scan unit that reads the image data one line at a time, a continuous pixel detection unit that detects the start point and the end point of the continuous black pixel in one line, and a section information storage unit that stores the start point and end point section information. A circumscribed rectangular frame determining unit that determines whether or not black pixels are connected by comparing the section information with the previous line information for each line and determines a circumscribed rectangular frame of the pattern; And a pattern identification unit that identifies the pattern shape by identifying the ratio of the number of black pixels to the number of white pixels.

[0008]

According to the present invention, the area of the circumscribed rectangular frame of the pattern is determined, and the number of black and white pixels in this frame is counted to obtain 2
Since it is possible to identify which one of the two patterns is used, a standard pattern is not required, there is no pattern matching processing, and the processing speed is high.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A method for identifying a specific pattern according to an embodiment of the present invention will be described below with reference to the drawings.

The identification of a left parenthesis "and a right parenthesis" indicating a range designation in a plate making machine will be described using a specific pattern as an example.

FIG. 1 is a block diagram showing an example of the overall configuration of the identification of a specific pattern according to the present invention. Reference numeral 1 denotes a pixel memory unit in which a specific pattern to be identified is stored.
Reference numeral 2 denotes a raster scanning unit which reads a pattern stored line by line from the pixel memory unit 1. The black pixel continuous section detection section 3 detects the start point and end point of the continuous black pixel in one line, and stores the section of the start point and end point in the section information storage section 4 as section information. Circumscribed rectangular frame determination unit 5
Then, the section information stored in the section information storage unit 4 is set to 1
For each line, it is compared with the section information one line before to determine whether or not there is a connection relationship, and a circumscribed rectangular frame of the specific pattern is determined. The pattern identification unit 6 identifies a pattern based on the quantity ratio between black pixels (part of a specific pattern) and white pixels in an area where the pixel memory is limited by a circumscribed rectangular frame.

FIG. 2 is an explanatory diagram for explaining the operation of the continuous section detection section 3 and section information storage section 4. 7, 8 are (i-1) th lines for one line of the pixel memory,
This is the pixel data of the i-th line. Reference numerals 9 and 10 denote storage locations of the (i−1) -th line and the i-th line in the section information storage unit 4. (I-
1) The start point and the end point of the section 11 and the section 12 of the second line are stored in the start point address 14 and the end point address 15, respectively. The start point and the end point of the section 13 of the i-th line are stored in the start point address 16 and the end point address 17, respectively.

FIG. 3 is an explanatory diagram for explaining the operation of the circumscribed rectangular frame determination unit 5. 18 is data in the pixel memory, and 1
9, 20, 21 are the second line, the third line, 4
This is a continuous section of black pixels on the second line. In determining the circumscribed rectangular frame, first, the connection relationship between the data of the n-th line and the data of the (n-1) -th line one line before is examined. Here, the connection relationship is, as shown in FIG.
-1) The X coordinate of the start point and end point of the line is a
and _n-1, b _n-1, the start point of the n-th line, the respective X-coordinate of the end point and a _n, b _n, if the (n-1) following relationship, black n th line It is assumed that pixel data is linked.

[0014] In the case of FIG. 3, a continuous section of black pixels starts from the second line, and at this stage, the circumscribed rectangular frame satisfies 5 ≦ X ≦ 7 (x direction) and 2 ≦ Y ≦ 2 (y direction). In the third line, the continuous section 20 of black pixels and the continuous section 19 of black pixels are connected, and the circumscribed rectangular frame is expanded to 3 ≦ X ≦ 7, 2 ≦ Y ≦ 3. In the fourth line, the circumscribed rectangular frame is also 3
≦ X ≦ 9, 2 ≦ Y ≦ 4. 21 on the 5th line
Since there is no continuous section of black pixels connected to the pattern, the circumscribed rectangular frame of this pattern is finally determined as 3 ≦ X ≦ 9, 2 ≦ Y ≦ 4.

FIGS. 5A and 5B show a pattern identifying section 6.
FIG. 4 is an explanatory diagram for explaining the operation of the present invention. 22 indicates a specific pattern, and 23 indicates a circumscribed rectangular frame. Here, circumscribed rectangular frame 2
The number of black pixels in the area of No. 3 and the number of white pixels in the area to the left of the black pixel are counted. For example, left parenthesis “and right parenthesis” is determined based on the pixel ratio as follows.

2 × (the number of white pixels on the left) <the total number of pixels in the circumscribed rectangular frame−the right parenthesis when the number of black pixels is “2 × (the number of white dots on the left)> the total number of dots in the outer frame− When the number of black dots is left parenthesis is determined.

Therefore, in the case of FIG. 5A, 2 × 9 <6
4-17, and is determined as “, and in the case of FIG.
× 49> 64−18, and is determined.

As described above, according to the present embodiment, the feature extraction of left parenthesis or right parenthesis is performed according to the ratio of the number of white and black dots of the pixels in the circumscribed rectangular frame. do not need. In this embodiment, since only the area of the circumscribed rectangular frame of “,” is determined and the number of black and white pixels is counted, the character normalization process and the rotation process are not required. The processing speed is fast without the need for matching processing.

In this embodiment, (,), <,
>, ｛,｝, Etc.

[0020]

As described above, according to the present invention, the area of a circumscribed rectangular frame of a pattern is determined, and only by counting the number of black and white pixels in this frame, it is possible to determine which of the two patterns. Since the identification can be performed, the processing speed is high without the need for the standard pattern and without the pattern matching processing.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an entire configuration of a specific pattern identification method according to an embodiment of the present invention;

FIG. 2 is an explanatory diagram for explaining functions of a continuous section detecting section and a section information storing section for black pixels according to the embodiment;

FIG. 3 is an explanatory diagram illustrating an operation of a circumscribed rectangular frame determination unit according to the embodiment;

FIG. 4 is an explanatory diagram showing a connection relationship of black pixels in the pattern of the embodiment.

FIGS. 5A and 5B are explanatory diagrams of a pattern identification unit for judging right parenthesis and left parenthesis of the embodiment, respectively.

[Explanation of symbols]

 Reference Signs List 1 pixel memory section 2 raster scan section 3 continuous section detection section for black pixels 4 section information storage section 5 circumscribed rectangular frame determination section 6 pattern identification section

Claims (1)

(57) [Claims] 1. A pixel memory unit for storing a pattern to be identified for identifying one of two patterns, and a raster for reading the pattern stored in the pixel memory unit line by line. A scanning unit, a black pixel continuous section detecting unit that detects a start point and an end point of a continuous black pixel in one line, a section information storage unit that stores a start point and an end point section as section information, and the section information. A circumscribed rectangular frame determining unit that determines whether or not black pixels are connected by comparing each line with the section information one line before, and determines a circumscribed rectangular frame of the pattern; A pattern identification unit comprising: a pattern identification unit that identifies a pattern shape by identifying a quantity ratio of pixels and determines which pattern is used.