JPS592069B2 - Graphical flaw detection device using contour line information - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a figure flaw detection device using contour line information, in which special symbols are attached to the contour line with a certain width on the top, bottom, left and right, and this figure is marked with the same width as that of the input figure to be inspected. The outlines are overlapped and the parts where they do not overlap are detected as graphic flaws.

When detecting graphic flaws by the matching method, conventional methods have poor flaw detection performance because a part that is not a flaw may be determined to be a flaw due to a very slight positional shift of the figure to be inspected.

Taking these drawbacks into consideration, the present invention accurately detects graphic damage even if there is a slight positional shift, depending on whether the contour line of the figure to be inspected falls within certain sections of the standard figure's contour line, top, bottom, left and right. It is something to detect.

A device according to an embodiment of the present invention will be explained according to the drawings.

FIG. 1 is a block diagram of an embodiment, in which a figure input device to be inspected 1,
It is composed of a first image memory 2, a position alignment device 3, a standard figure input device 4, a second image memory 5, a boundary line extraction device 6, a boundary line section setting device 7, a flaw detection device 8, and a display device 9. There is. The figure input device 1 to be inspected is a film reader,
It consists of input devices such as a flyshield spot scanner and a TV (television) camera. Film or drawings are input, digital image data is converted, and the first
The image is stored in the image memory 2 of. At this time, the second image memory 5
The alignment device 3 aligns the standard figure and the input figure using the standard figure data in the input figure. The standard figure input device 4 consists of any input device such as a film reader, a flying spot scanner, a TV camera, etc., as well as a figure input device (digitizer) for standard figure design.
As a result, the standard figure is input, becomes digital image data, and is stored in the second image memory 5. The boundary line extraction device 6 extracts the boundary lines of figures stored in the first or second image memory 2 or 5. The boundary line section setting device 7 attaches corresponding symbols or numerical values to sections of a certain width on the upper, lower, left, and right sides of the boundary line, and stores them in the second image memory 5. Scratch detection device 8
detects graphic flaws by superimposing the data in the first image memory 2 and the data in the second image memory 5 and outputs it to a display device 9 such as a CRT display device.
Next, FIGS. 2 and 3 are a standard figure line diagram and a boundary line diagram of one of the objects to be processed when implementing the present invention, FIG. 4 is a figure line diagram in which the boundaries of the standard figure are set, and FIG. The figure is a table showing the digital data of the boundary line, Figure 6 is a table showing sections set on the boundary line of Figure 5, Figure 7 is a figure line diagram of the object to be inspected, and Figure 8 is the figure 7 Extracted boundary diagram of figures in Figure 9
The figure is a line diagram of the boundary line section area in a certain direction of the area where the boundary line section is set in Fig. 4, and Fig. 10 is a diagram of the boundary line of the figure to be inspected in Fig. 8 and the boundary in a certain direction in Fig. 9. FIG. 11, which is a diagram in which the line section areas are superimposed, is a diagram showing the results of these inspections.

Next, centering on the block diagram in Figure 1, we will move on to Figures 2 to 11.
The graphic flaw detection device will be explained according to the figures.

First, the standard figure shown in FIG. 2 is inputted from the standard figure input device 4 and stored in the second image memory 5, and then the boundary line is extracted by the boundary line extraction device 6 as shown in FIG. As shown in FIG. 4, the section setting device 7 enters corresponding numerical values into sections of a certain width on the left and right sides of the boundary line P, and stores them in the second image memory 5. In FIG. 4, C is the upper area of the boundary line P, D is the lower area, E is the left area, and F is the right area. As an example of how to enter this value, we will explain a case where an area is set with a width of "1" for both the top, bottom, left, and right.The border P part is now "1", and the other parts are [0].
Then, an example of real image data is as shown in Figure 5.
Expressed in coordinates X, y, Xly4, X2y3, X3)Y
32X4y3X5yl2X5y2X5y3 is "1"
It is.

Next, if we take the boundary line image space as shown in Figure 12 and take I on the horizontal axis and J on the vertical axis, the boundary line image data becomes F (1, J) (1≦
I≦a;≦J≦b1 (where A and b are natural numbers), and I-1. ．． The boundary line image data of Jj is divided into portions Q as F(1,j). Furthermore, the 3×3 subspace of the (1,j) image space is shown in FIG. 13, and m−1 in the horizontal I direction.
, Mlm+1, and n-1, Nln+1 in the vertical J direction.
, then data unit F(K,l) (where m-1≦k≦
m+1:n-1≦l≦n+1), F(RIl±1,
. ±1) is arranged as shown in the figure and is changed by F(.±1, .±1). That is, (P, q, r, S)-F(M,
n-1) to F(m+1,n)) If F(m-1,.) is F(.N,.)-0, F(N.,n) is changed as shown in the table below. Ru. Here, -2,2 is the left and right direction,
4 and -4 represent boundary line section areas in the vertical direction, and 3, -3, 5, and -5 represent boundary line section areas in the 45 direction.

By performing this changing operation over m=1 to A and n-1 to b, the boundary line section setting operation is completed as shown in FIG. However, as shown in the boundary line image data in Figure 15, F
The end of (1, J) is surrounded by "O". In addition, to add an explanation of the contents of the above-mentioned table, the first table in the table is Figure 14 a, the second is Figure 14 b, the third is Figure 14 c, the seventh is Figure 14 D, and the ninth is Figure 14. The first one appears in e of the same figure, and the eleventh one appears in f of the same figure. Also in this case, X2y2, X3y4, X4y4
,X5y5,Xly5 is -2)X2y3FX3y3'X
4y32X5yl2X5y29X5y3 is 1)Xly3
9X4y2 is 3, X6yl,X6y2tsuX6y3 is -4
).

Next, the figure to be inspected shown in FIG. 7 is inputted from the figure to be inspected input device 1 and stored in the first image memory 2, and then the boundary line is extracted by the boundary line extraction device 6 as shown in FIG. , the original image data of the first image memory 2 and the image data of the second image memory 5 shown in FIG. When inputting, if the direction of slight positional deviation from the standard figure is known in advance, for example, if the upper area C and left area F are shifted upward and leftward with respect to the boundary line P, the second From the image data of FIG. 4 in which the boundary line section is set in the image memory 5, only the boundary line section in the corresponding direction (4
By extracting the boundary line section in Fig. 9 (including the area in the 5-direction) and overlapping the boundary line of this figure and the inspected figure in Fig. 8 as shown in Fig. 10, the solid line section in Fig. 11 is obtained. Figure flaws in the figure to be inspected as shown by A and B are detected and output to the display device 9. If the direction of slight misalignment with the standard shape when inputting the figure to be inspected is not clearly known, check the overlap between the boundary line of the figure to be inspected and the figure that has all the areas in the area where the boundary line section has been set. The alignment is performed to detect and display graphic flaws. Thus, with the device of the present invention, it is possible to accurately detect a figure flaw even if there is a slight positional deviation, depending on whether or not the outline of the figure to be inspected falls within a certain section on the left and right sides of the contour of the standard figure. It is.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG.
Fig. 3 is a diagram of one standard figure and boundary line to be processed when implementing the present invention, Fig. 4 is a figure line diagram with sections set for the boundary line of the standard figure, and Fig. 5 is digital data of the boundary line. Figure 6 is a table diagram showing the boundaries of Figure 5, Figure 7 is a line diagram of the figure to be inspected, and Figure 8 is a diagram of the figure extracted from Figure 7. Boundary line diagram, Figure 9 is a line diagram of the boundary line section area in a certain direction of the area where the boundary line section is set in Figure 4, and Figure 10 is the boundary line of the figure to be inspected in Figure 8 and Figure 9. FIG. 11 is a diagram showing the results of these inspections. 12 and 15 show boundary line image data diagrams, and FIGS. 13 and 14 a to 14 show 3×3 partial space diagrams of different image spaces. 1 is a graphic input device to be inspected, 2 is a first image memory, 3 is a positioning device, 4 is a standard graphic input device, 5
1 is a second image memory, 6 is a boundary line extraction device, 7 is a boundary line section setting device, 8 is a flaw detection device, and 9 is a display device.

Claims (1)

[Scope of Claims] 1. A graphic input device to be inspected that inputs film or drawings and converts them into digital image data, a first memory that stores image data from the input device, a standard graphic input device, and a standard graphic input device. a second image memory that stores standard figure data as digital image data; a positioning device that performs alignment with an input figure using the standard figure data in the second image memory; a boundary line extracting device for extracting a boundary line of a figure stored in an image memory; a boundary line extraction device for extracting a boundary line of a figure stored in an image memory; a boundary line extraction device for extracting a boundary line of a figure stored in the second image memory; a boundary line section setting device for storing, a graphic flaw detection device for detecting graphic flaws by superimposing data in the first image memory and data in the second image memory, a display device for displaying the graphic flaw detection section; A graphic flaw detection device using contour information, comprising: 2 If the direction of slight positional deviation from the standard figure is known when the figure to be inspected is input, the boundary line in the corresponding direction is determined from the standard image data set in the boundary line section in the second image memory. The contour line information according to claim 1 is characterized in that a graphic flaw in the figure to be inspected is detected by extracting a boundary line section consisting only of sections and overlapping the boundary line of this figure and the figure to be inspected. The graphical flaw detection device used.