JPS617406A - Object shape defect detection method - 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 [Technical Field of the Invention] The present invention relates to an object shape defect detection method for detecting shape defects of an object having a predetermined shape by image processing.

[Prior art]

FIG. 1 is a block diagram showing the conventional configuration of a defect detection device using this type of defect detection method, in which 1 indicates an object to be inspected;
2 is a TV camera, 3 is an AD converter, 4 is an image memory,
5 is an image memory storing the image signal of the reference object, 6 is a comparison circuit, and 7 is a CP consisting of a microphone o7' processor, etc.
U, 8 is a determination circuit, and 9 is an output terminal.

In such a configuration, an image of the object to be inspected l is captured by the television camera 2. Then, the image signal is converted into image data of, for example, 16 gradations by the AD converter 3 and stored in the image memory 4.

As a result, image data representing the density of the image of the object to be inspected 1 is stored in the image memory 4 as codes from "0" to "F" (in hexadecimal notation) according to the density, as shown in FIG. 2a. Ru.

On the other hand, image data representing the density of the image of the reference object 2o is stored in the image data memory 5 as "0" as in the image memory 4.
~"F' (see Fig. 2). When the image data of the object to be inspected 1 is stored in the image memory 4, this image data is read out under the read control of the CPU 7 and sent to the comparison circuit. 6. At the same time, the image memory 5
The image data of the reference object 20 stored in advance in the CP
The data is read out by the readout control of U7 and inputted to the comparison port 6, where it is compared with the image data of the object to be inspected 1.

As a result, if the two image data completely match, a signal indicating that there is no defect is output from the sidecar circuit 8. In this case, comparison of image data is performed pixel by pixel. Also,
In order to prevent erroneous judgments due to errors in adjusting the image signal, the judgment signal indicating the presence of a defect is determined when the image data of the object to be inspected 1 and the image data of the reference object 20 are equal to or greater than a predetermined value in the density region. It is designed to be output only if

Therefore, according to this configuration, only when the defective portion 3 shown in FIG.
The image data 33 of the pixel corresponding to the reference object 20 does not match the image data 21 of the reference object 20, and a determination signal indicating that there is a defect is output from the determination circuit 8.

However, according to such conventional defect detection methods, ■
Since the image data of the inspected object 1 and the pixel data of the reference object 20 are sequentially compared pixel by pixel, the presence or absence of defects can be detected with high precision, but the drawback is that it takes a long time to complete the determination. There is.

[Summary of the invention]

The present invention has been made to solve these drawbacks, and it averages the image signals of the object to be inspected and the reference object in units of a predetermined number of pixels in the density area, and compares the average values. The object of the present invention is to provide a method for detecting defects in the shape of an object, which can detect the presence or absence of defects in an object.

[Embodiments of the invention]

FIG. 3 is a block diagram showing an embodiment of a defect detection device to which the present invention is applied. The difference from the conventional configuration is that an average value calculation circuit 1° is provided on the output side of the image memory 4, and a
This is because the judgment port 98 in the figure has been deleted.

In such a configuration, image data of the object to be inspected 1 is stored in the image memory 4 in the same manner as in the prior art. On the other hand, as shown in FIG. 4a, the image memory 5 contains averaged image data of "0" obtained by averaging the image data of the reference object 20 in units of a predetermined number of pixels (in units of 4 pixels in this embodiment) in the density area. ~
It is stored in advance with the code "F".

Therefore, when the image of the object to be inspected 1 is stored in the image memory 4, this image data is read out under the readout control of the CPU 7 and input to the average value calculation circuit 10. and,
This average value calculation circuit 10 calculates the average value of the density in units of four pixels.

FIG. 4 shows image data in this case.

In parallel with this operation, the averaged image data of the reference object 20 is read from the image memory 5 in units of four pixels, and is compared with the average value calculated by the calculation circuit 10 in the comparator 6.

As a result of this comparison, if the two averaged image data match, a determination signal indicating no defect is output from the comparison circuit 6. The comparison operation in this case is performed in units of four pixels.

Therefore, the determination result of this embodiment is output in the conventional required time of x. In this case, the time required can be further reduced by increasing the number of pixels to be averaged, but
On the other hand, the accuracy decreases, so it is desirable to set the number of averaging pixels to an appropriate number in relation to the accuracy.

〔Effect of the invention〕

Since the present invention is designed to detect the presence or absence of defects, it is possible to detect the presence or absence of defects in the object to be inspected in an extremely short period of time, and when applied to a defect detection device in a production line, the efficiency of the production line can be improved.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of a conventional defect detection bag device, Fig. 2 is an image data array diagram for explaining the operation of the conventional device, and Fig. 3 is an example of a defect detection device to which the present invention is applied. A block diagram showing an embodiment, FIG. 4 is an array diagram of averaged image data. 1... object to be inspected, 2... television camera, 3...
AD converter, 4, 5... Image memory, 6... Comparison circuit, 7... CPU, 31... Defective part. Agent: Masuo Oiwa (2 others) Figure 1 name 3 Figure 4 G) 4th b)

Claims (1)

[Claims] In an object shape defect detection method in which a defect in the shape of the object to be inspected is detected by extracting an image of the object to be inspected using an imaging device and comparing it with an image of a reference object, the object to be inspected extracted by the imaging device and the reference 1. A method for detecting defects in the shape of an object, characterized in that image signals of the object are averaged in units of a predetermined number of pixels in a density region, and shape defects in the object to be inspected are detected by comparing the averaged values.