JP2882314B2 - Line figure recognition method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a binary two-dimensional digital image obtained by inputting a map, drawing, flow chart, or the like drawn on a paper surface with a line figure using an image scanner or the like. The present invention relates to a line graphic recognition method capable of detecting a closed figure and recognizing attributes such as the type even if the line is broken or the type of line is a broken line or a dashed line.

This line figure recognition method can be widely used as a basic technique for realizing automatic reading of maps and drawings.

[0003]

2. Description of the Related Art Conventionally, a binary two-dimensional digital image I in which a graphic portion is represented by a pixel having a value of 1 and a background portion is represented by a pixel having a value of 0
The following method is used to detect a closed figure and recognize its attribute.

First, a thinning process is performed on the image I to generate a thinned image I '. The thinning process is a process of repeatedly performing an operation of converting a pixel included in a graphic portion and in contact with the background into 0 during raster scanning of the image until the graphic becomes a core line graphic having a width of 1.

Next, the cored figure on the thinned image I 'is traced and approximated by a series of coordinate values. Thereafter, connection processing of a gap having a short distance is performed in consideration of detection of a closed figure with a break and a closed figure drawn by a broken line and a dashed line. The connection process is a process of selecting a combination of a pair of open end points that has a sufficiently small distance from each other and that cannot form a corner with a large angle when connected, and interpolating a gap between these open end points and connecting. is there. The distance between the open end points and the threshold value of the corner angle for determining whether or not to make the connection are given in advance.

Next, a part forming a closed loop is detected on the series data of coordinate values obtained as a result of the gap connection processing.

Next, approximation processing is performed on a series of coordinate values forming a closed loop in order to remove the influence of a digitization error. For the closed loop subjected to the approximation processing, the shape determination is performed by measuring the number of vertices, the size of the interior angle, the length of the side, the parallelism of the facing side, and the like. The shape determination result is used as the recognition result of the closed figure corresponding to the closed loop.

[0008]

However, in the conventional method of detecting a closed figure after interpolating a break in a line figure after interpolating a break in a line figure, a connection candidate for a plurality of end point pairs can be seen in the interpolation connection processing as shown in FIG. It can be difficult to detect the correct endpoint pair to connect when is detected. As a result, there has been a problem that a closed figure cannot be detected or an incorrect closed figure is detected.

[0009]

SUMMARY OF THE INVENTION The present invention relates to a line figure recognition system for recognizing a closed figure drawn by a closed line on a binary two-dimensional digital image. An initial polygon generating unit that generates a polygon inside a closed graphic on a three-dimensional digital image, a polygon fitting unit that receives the polygon and the two-dimensional digital image as input and expands and deforms the polygon until it inscribes the closed graphic, result of expansion deformation, there a line figure recognition method and a determining figure detection unit attributes such as the type of the closed figure by measuring the shape characteristics of the polygon the polygon inscribed in the closed figures as input
The initial polygon generating unit may
Graphic expansion processing means for generating an image,
Inner contour detecting means for detecting the contour, and
Polygon with similar processing and predetermined N nodes
And a polygon generating means for generating a polygon .

[0010]

The operation of the present invention will be described with reference to FIG.

When a binary image I including a closed figure is input to the initial polygon generating unit 11, the initial polygon generating unit 11 first generates an image G obtained by performing a graphic expansion process on the input image I. Here, the graphic expansion processing determines whether or not a background pixel adjacent to a contour pixel is determined based on the arrangement of pixel values in a 3 × 3 pixel area centered on a point of interest while raster-scanning the image. Is a process of rewriting the pixel value from 0 to 1 if the pixel is a background pixel adjacent to. The contour pixels are pixels included in the figure portion and in contact with the background. In one process, the figure expands toward the background by one pixel. The image G is an image that has been subjected to about 1/2 of the gap distance to be connected with the graphic expansion processing, and the gap between the line segments is in a filled state.

Next, of the outlines of each figure on the image G, an inner outline existing inside the figure is detected. Further, the detected inner contour is approximated, and the obtained closed coordinate value series is output as an initial polygon.

Next, in the polygon fitting section 12, the initial polygon is expanded and deformed on the input image I to inscribe a closed figure on the image. A method of expanding and deforming the initial polygon will be described in an embodiment.

Next, the figure detection unit 13 determines the shape by measuring the number of vertices of the polygon obtained as a result of the expansion deformation, the size of the interior angle, the length of the side, the parallelism of the opposite side, and the like. The shape determination result of the polygon is used as the determination result of the closed figure with which the polygon is in contact.

As described above, first, a polygon is generated in a closed figure, and then the polygon is expanded and deformed and inscribed in the closed figure. It becomes possible to detect the shape of a figure. As a result, the problem that a correct polygon is not detected due to an error in the conventional connection processing can be solved.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the overall configuration of the present invention will be described. FIG. 2 is a configuration diagram of one embodiment of the present invention, and the same parts as those in FIG. 1 are denoted by the same reference numerals.

Reference numeral 11 denotes an initial polygon generating unit for generating and outputting a closed figure drawn by a solid line, a broken line, an alternate long and short dash line, or a closed figure drawn by a solid line on an input binary image. It is. The initial polygon is a polygon formed by sequentially connecting N nodes Q1, Q2,..., QN. Here, the number of nodes N is given in advance. The initial polygon generation unit 11 includes a figure expansion processing unit 11a, an inner contour extraction unit 11b, and a polygon generation unit 11c.

The graphic expansion processing section 11a performs expansion processing of the input binary image I a predetermined number of times. The image I input to the graphic expansion processing unit 11a is first raster-scanned, and a mark is given to a pixel adjacent to the contour pixel. Here, the mark is added to a plane different from the plane that stores the pixel values of the image I. The determination as to whether or not the pixel is adjacent to the contour pixel is made based on the arrangement of pixel values in a 3 × 3 pixel area centered on the target pixel during raster scanning.
When the pixel value of the target pixel is 0 and any one of the eight pixels in the vicinity is 1, the target pixel is determined to be the background pixel to be obtained. After the raster scan of the entire image I is completed, the value of the marked pixel is rewritten from 0 to 1. Through the above processing, an image G in which the figure is expanded by one pixel toward the background is generated.

The inner contour extraction unit 11b extracts the outline of the black pixel connected component in the input binary image G by tracing the outline, and then selects and outputs only the inner outline. The contour tracking is a process of detecting contour pixels of a black pixel connected component that has not been found by raster scanning, and subsequently tracking and detecting contour pixels adjacent to the contour pixel on the black pixel connected component one after another. The tracked pixel is given a tracked mark on another plane of the image G. When a contour for one round is detected, raster scanning is performed again, and when an undetected contour pixel is detected, the same processing as described above is performed. As a result of the above processing, contours in all images are detected. The contour detection result is stored as a sequence of coordinate values of the contour pixels.

The detection of contour pixels by raster scanning is performed by examining the value of the pixel of interest p and its neighboring pixels {p0, p2, p4, p6}. FIG. 3 shows the relative positional relationship between the target pixel p and its neighboring pixels {p0, p2, p4, p6}. The value of the pixel of interest p is 1, and the neighboring pixels of {p0, p
2, p4, p6} has a value of 0, the pixel of interest p
Is a contour pixel.

For each contour detected by raster scanning and contour pixel tracking processing, the sum of declination when the series of coordinate values is regarded as a polygon is calculated, and the outer contour of the figure is replaced with the inner contour ( (Inner contour). The sum of declination is +2
When it becomes π, it is an outer contour, and when it becomes -2π, it is an inner contour. The series of each coordinate value of the inner contour of the figure detected by the above processing is output to the polygon generation unit 11c.

The polygon generating unit 11c is a process for inputting a series of coordinate values of the inner contour and approximating the inner contour with a polygon having a predetermined number N of vertices. The input coordinate value series of each contour is subjected to a thinning process to be converted into a polygon having N vertices.

The configuration of the initial polygon generator 11 has been described above.

Reference numeral 12 denotes a polygon generated by the initial polygon generation unit 11 which is expanded and deformed on the input image I to make it inscribed in a closed figure, and outputs the polygon inscribed to the closed figure to the figure detection unit 13. This is a polygon fitting section. The polygon fitting unit 12 includes a polygon deformation unit 12a and a deformation stop determination unit 12b.

The polygon deformation unit 12a performs expansion deformation on the input polygon. The polygon is input from the initial polygon generation unit 11 or the deformation stop determination unit 12b. In the expansion deformation, for three consecutive nodes Qn-1, Qn, and Qn + 1 among the nodes Q1, Q2,. At this time, the positions of Qn-1 and Qn + 1 are moved by a small distance in a direction to widen the interior angle. However, with respect to the polygon re-input from the deformation stop determination unit 12b, the movement processing is performed only on the permitted nodes.
The above operation is performed for all combinations of three consecutive nodes. The polygon after the expansion deformation is the deformation stop determination unit 1
2b.

The deformation stop judging unit 12b calculates the sum of the distance between the input polygon and the figure containing the polygon and the sum of the interior angles, and calculates the nodes Q1, Q2,.
-It is determined whether or not further movement of the QN is permitted. If the distance dn between the node Qn and the figure is dn> 0, further movement of the node Qn is allowed. If there are no more deformable nodes, the polygon is determined to be inscribed in the closed figure, and the dilation process is completed. The polygon for which the expansion processing has been completed is output to the graphic detection unit 13.

The configuration of the polygon fitting unit 12 has been described above.

Reference numeral 13 denotes a polygon output from the polygon fitting unit, which is first approximated, and then the number of corners, the corner angle, the length of the side, the parallelism of the facing side, and the like are measured. This is a graphic determining unit that determines the attribute of the closed graphic and outputs the final result. The type of polygon is determined based on the number of corners, and the type of a special polygon such as a parallelogram or a square is determined based on the parallelism of the facing sides and the length of the sides. Conditions necessary for these attribute determinations are registered in advance as dictionary data.

The above is the overall configuration.

Next, the overall operation of the present invention will be described.

When a binary image I including a closed figure is input to the initial polygon detection unit 11, a figure expansion processing is performed by a figure expansion processing unit 11a to generate an image G in which gaps such as breaks and broken lines are filled. You. At this time, it is assumed that the number of times of the graphic expansion processing is set in advance to about の of the gap distance to be connected. The image G is the inner contour detection unit 11
b.

The inner contour detector 11b performs a process of detecting an inner contour in the image G, and outputs a result of the detection. The detection result of the inner contour is input to the polygon generation unit 11c, where it is approximated and converted into a polygon having predetermined N nodes.
Output to This polygon is called an initial polygon. The initial polygon and the binary image I are input to the polygon fitting unit 12.

The polygon deforming section 12a inside the polygon fitting section 12 expands and deforms the input initial polygon by a very small amount. Using the binary image I, it is determined whether the polygon expanded and deformed by the minute amount is inscribed in the closed figure using the binary image I. If it is determined that the polygon is not inscribed, the polygon evaluated to the polygon deforming unit is re-input together with information on a node that can be further moved. At this time, the polygon deformation unit 12a further performs expansion deformation on the re-input polygon. As a result of repetition of the above processing, when the deformation stop determination unit 12 b determines that the polygon is inscribed in the closed figure, the polygon is output to the figure detection unit 13. FIGS. 4A and 4B show examples of the initial polygon and the final polygon output to the graphic detection unit 13, respectively.
And (b).

Next, in the figure detecting section 13, the polygon obtained as a result of the expansion deformation is approximated. The number of vertices, the size of the interior angle, the length of the side, the parallelism of the facing side, and the like are measured for the approximated polygon, and the shape of the closed figure in which the polygon is inscribed is determined.

[0035]

As described above, according to the present invention, gaps between line elements such as broken lines and the like caused by the discontinuity of closed figures and the closed figure being drawn by broken lines are different from the prior art. Rather than performing interpolation connection from open end point search, the polygon generated inside the closed figure on the input image is expanded and deformed until it is inscribed in the closed figure, and the attribute of the closed figure is determined based on the shape of the obtained polygon. It was configured as follows. as a result,
In the conventional method, it is possible to correctly detect and recognize the attribute of a closed figure which could not be detected because a break point cannot be uniquely determined. Therefore, it is possible to perform highly accurate closed figure recognition that exceeds the conventional method.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of the present invention.

FIG. 2 is a block diagram showing one embodiment of the present invention.

FIG. 3 is a diagram for explaining a method of detecting a contour pixel;

4A is a diagram showing a relationship between a closed figure and an initial polygon, and FIG. 4B is a diagram showing a result of expanding and deforming the initial polygon.

FIG. 5 is a diagram showing an example of an image in which it is difficult to detect a closed figure by a conventional method.

[Explanation of symbols]

 Reference Signs List 11 Initial polygon generation unit 11a Graphic expansion processing unit 11b Inside contour extraction unit 11c Polygon generation unit 12 Polygon fitting unit 12a Polygon deformation unit 12b Deformation stop determination unit 13 Graphic detection unit

──────────────────────────────────────────────────続 き Continued on the front page (58) Fields surveyed (Int.Cl. ⁶ , DB name) G06T 7/00 G06T 9/20 JICST file (JOIS)

Claims (3)

(57) [Claims] A line figure recognition system for recognizing a closed figure drawn by a closed line on a binary two-dimensional digital image,
An initial polygon generation unit that receives the two-dimensional digital image as input and generates a polygon inside a closed graphic on the two-dimensional digital image, and receives the polygon and the two-dimensional digital image as input and inscribes the closed graphic. A polygon fitting portion for expanding and deforming a polygon, and a graphic detection for determining attributes such as a type of the closed graphic by measuring graphic characteristics of the polygon by inputting the polygon inscribed in the closed graphic as a result of the expansion and deformation. A line figure recognition method, comprising:
Is a figure dilation process that generates an image with a gap filled
Means and inner contour detection for detecting an inner contour in the image
Means for approximating the inner contour and determining a predetermined N
Polygon generating means for generating a polygon having a plurality of nodes
And a line figure recognition method. 2. The apparatus according to claim 1, wherein said polygon fitting portion is
Polygon deformation means for expanding and deforming the initial polygon by a small amount
And the polygon expanded and deformed by the minute amount is inscribed in a closed figure.
And a deformation stop determining means for determining whether to perform
2. The line graphic recognition method according to claim 1, wherein: 3. The method according to claim 1, wherein the figure detecting section obtains a result of the expansion and deformation.
Approximated polygons and vertices for the approximated polygons
Number, interior angle size, side length, parallelism of opposing sides, etc.
And determine the shape of the closed figure in which the polygon is inscribed
3. A method for recognizing a line figure according to claim 1, wherein
formula.