JPH03246776A - Recognizing method for outside shape of object - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for recognizing the external shape of an article by screen processing,
It is particularly effective for recognizing the external shape of electronic components in the process of mounting electronic components on printed circuit boards.

[Prior Art] One of the important steps in surface mounting electronic components onto a printed circuit board is the step of accurately mounting the electronic component to be mounted at a predetermined position on the printed circuit board. In order to carry out accurate mounting numbers, it is necessary to accurately and without error detect the external shape of the electronic components to be mounted, and various detection and recognition methods are used.

One of the effective recognition methods is a method using image processing as described below.

To explain this with reference to FIG. 3, an image 1 obtained by photographing a sample such as an electronic component using an optical device (not shown) is displayed in various shades of density.

However, with this method, it is not possible to clearly distinguish the shape of the sample, so the entire area of the screen 3 displayed in various shades of density is determined by whether it is darker or lighter than one predetermined density reference value T. It is converted into a value and displayed in an image area divided into two colors, white and black. In this way, the image 2 is binarized and displayed in two colors. In this image 2, a portion 2a (indicated by “1”) darker than the gradation reference value T is shown in black, and other portions are shown in the same color as the screen background area 3' of the screen 3.

Incidentally, since the unevenness on the surface of the sample appears in the shading of the image, the unevenness on the surface as shown in FIGS.
When the above-described processing is performed on the sample 4 having the concave portion 4c, an image 5 as shown in FIG. 5 is obtained.

[Problems to be Solved by the Invention] In a method for recognizing the outer shape of a sample using such a conventional image processing method, in the image 5 of the sample 4, an image area corresponding to the outer frame portion 4a, which is a high part of the surface of the sample 4; 5a and the image area 5b corresponding to the symbol portion 4b have a complementary color to the screen background area 3' of the screen 3, but the image area 5c corresponding to the other lower part of the surface of the article 4 is the screen background area of the screen 3. Since it is the same color as 3', the symbol on the surface of the article and the outer periphery 4a
This has the disadvantage that even the inner edge of the article 4 is recognized as an image, making it easy to misrecognize the outer shape of the article 4. In other words, with such conventional image processing methods, it is difficult to accurately recognize the outer shape of an object whose outer shape is to be recognized when the object has irregularities on its surface.

Therefore, conventional surface mount equipment has not yet been able to accurately recognize the external shape of certain types of electronic components, such as electronic components with uneven package surfaces. This has been one of the causes of lowering the reliability of mounting, and has also placed restrictions on the mounting of electronic components of such specific shapes.

[Means for Solving the Problems] In order to eliminate such conventional drawbacks, the present invention provides a method for recognizing the external shape of an object as set forth in the claims.

[Function] According to the present invention, even if the surface of the object to be recognized has various unevenness, the image of the object can be expressed in one color.
Therefore, according to the present invention, even if the object has various irregularities on its surface, its outer shape can be accurately recognized.

[Example of implementation] First, the basic concept of external shape recognition by image processing according to the present invention will be described. It is not a screen background area, but an image area that has a complementary color relationship to the color scheme of this image background area, and the outer circumference of the outermost image area corresponds to the outline of the object that is the subject of image processing. Image processing is performed based on this concept.

Next, an embodiment of the present invention will be described with reference to FIGS. 1 and 2.

The object to be subjected to image processing is, for example, an electronic component 4 having an outer frame portion 4a having a substantially constant height (i.e., level) on its surface, as shown in FIG. 4, for example. Here, the outer frame does not need to be clearly bordered as shown in Figure 4, but it should be such that the substantial outermost periphery of the surface of the object to be image processed is at approximately a constant height. means.

(1) M-image photography First, as in the conventional method described above, an original image of the electronic component 4 whose external shape is to be accurately recognized is photographed.

(2) Binarization screen processing The captured image is then binarized and screen processed to obtain a black and white processed screen 3 as shown in FIG. 2(a). This screen 3 includes a screen background area 3'', an image area 5a corresponding to the outer frame 4a of the sample 4, which has a complementary color to the color scheme of the screen background area 3', and an image area 5b corresponding to the symbol 4b. This area corresponds to a low area other than the screen background area 3' and is composed of an image area 5c of the same color as the screen background area 3'.

(3) Area calculation same color processing Next, area calculation and color matching processing, which has not been performed conventionally, is performed.

This process sequentially makes the image areas located inside the sample image 5 the same color except for the image area 5a corresponding to the outer frame 4a, and finally makes the image areas the same color as the outermost image background area 3'. In the image area inside the image area 5a corresponding to the outer frame part 4a in the sample image 5, the areas of the image area 5b and the image area 5C are compared and calculated to determine the size of the area. Then, the image area 5b with a small area is changed to the color scheme of the image area 5c with a large area to make it uniform.

This process only needs to be carried out once in the case of an electronic component as shown in FIG. ), it is performed multiple times.

(4) Screen black-and-white inversion processing After that, the color arrangement of the one-screen background area 3' and the image area 5a corresponding to the outer frame portion having a complementary color relationship (white) is inverted. As a result, a screen 3 having a color scheme as shown in FIG. 2(b) is obtained.

The electronic component image 6 on the screen 3 consists of a black image area 6a corresponding to the outer frame part 4a of the sample 4 and a white image area 6b corresponding to the inner part thereof.

(5) Image coloring process Next, the inner image area 6b of the inverted electronic component image 6 in FIG. 2(b) is filled with black to create a black coloring as shown in FIG. 2(C). A uniform filled electronic component image 7 is obtained.

−<6) Screen black and white reversal process Finally, by reversing the color scheme of screen 3, the second
The screen becomes a screen having a white re-inverted electronic component image 8 as shown in Figure (d). This process is a process for converting the external shape of the sample 4 clarified in FIG. 4(C) into a measurable color scheme.

Recognizing the color scheme of the re-inverted sample image 8 shown in FIG. 2(d),
By calculating the positions of, for example, the four corners of the re-inverted sample image 8, it is possible to measure the mounting accuracy of the electronic components mounted on the printed circuit board.

[Effects of the Invention] As described above, according to the present invention, even an object having an uneven pattern on its surface can reliably and accurately recognize its external shape. It is also possible to reliably and accurately detect the mounting position of an electronic component whose surface is marked with a frame or a symbol by concavo-convex surfaces. Furthermore, the measurement time for electronic components whose surfaces are marked with irregularities such as frames or symbols is reduced to about 115 compared to the conventional method, thereby improving the efficiency of inspecting the mounting position.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a diagram showing an image processing process for explaining an embodiment of the method for recognizing the outer shape of an object according to the present invention, and FIG. 2(a)
~ (d) are diagrams showing processed images in the image processing step shown in Figure 1, Figure 3 is a diagram for explaining a processing screen and its processing method by conventional image processing technology, and Figure 4 (A) ,
(B) is a diagram showing a sample subjected to image processing, and FIG. 5 is a diagram showing a processing screen according to the prior art. 1... Photographed image 3... Processing screen 5... Sample image 7... Filled sample image 2... Binarized processed image 4... Sample 6... Inverted sample image 8... Re-inverted sample image (1 other person) Figure ■ Figure (0) Shade distribution at the time of photography (×) Figure (0) Binarization of gray scale distribution (×)

Claims (1)

[Scope of Claims] 1. A method for recognizing the outer shape of an object having an outer frame portion having a substantially uniform height on the photographed surface, wherein the object is photographed to obtain a grayscale image thereof, and the grayscale image is A process of performing binarization processing and distributing two colors based on a predetermined reference density;
For the image expressed in two colors by the two-color distribution process, the area of each color of the image area existing inside the inner edge of the outer frame of the object is calculated, and the area within the image area is calculated. A process of unifying the color of a large area and filling it with one color, and changing the color of the image area filled with the one color to the color of an image area corresponding to the outer frame of the object, A method for recognizing the outer shape of an object, the method comprising: processing to make an image of one color; and 2. The method for recognizing the outer shape of an object according to claim 1, further comprising performing a process of reversing the color of the screen after the last process.