JPS60152937A - Defect extracting method in magnetic particle inspection - Google Patents

Abstract

PURPOSE:To make it possible to perform automatic judgment of defects and harmful pseudo-defects and automatic visual inspection based on the detection of the change in pictures, by performing absolute value difference processing of the pictures before and after the magnetic particle inspection of sample bodies by using an ITV camera and a picture processing device. CONSTITUTION:The pictures of a sample body 14 having defects and a sample body 17 having harmful pseudo-defects before an MT inspection are recorded in a picture memory 22. The recording before and after the MT inspection is controlled by a control signal 21. The pictures 15 and 18 after the inspection are compared in an operator 23, and a defect signal 24 indicated by E is obtained. In this way, the defects 16 and the harmful pseudo-defects 19 of the sample bodies 14 and 17 at the time of the MT inspection can be automatically judged by using the absolute value difference method of the pictures. By the introduction of the ITV camera and a picture processing device, automatic visual processing can be performed.

Description

[Detailed description of the invention] The present invention relates to a defect extraction method in magnetic particle inspection.

For example, as shown in Figure 1 (B), magnetic particle inspection (
The main objects to be inspected in the MT inspection (hereinafter referred to as MT inspection) are those having a welded part 2 like the test specimen 1. In such inspection objects, defects 3 often occur at the boundaries of the welded portions 2. Generally, undercuts 4, spatters 5, etc. are usually present in the vicinity of the welded portion 2 due to processing, and it is necessary to distinguish between the defects 3 and the undercuts 4, spatters 5, etc. in the MT inspection.

As shown in Fig. 2) and (B), due to the demand for automation of MT inspection, during the MT inspection of the test specimen 1, magnetic particle scattering 7 is carried out at the same time as excitation by the magnetic particle flaw detector 6 as shown in Fig. 2 (g). After removing unnecessary magnetic particles with air blower 8, the magnetic particle pattern of defect 3 that remains as shown in Figure 2 (B) is examined by IT.
The possibility of automatic identification through observation using a V-camera 9 and illumination 10 has been studied. However, as shown in FIGS. 3) and 3(B), the ITV camera scanning signal 11 in the test specimen 1
．． 12.11' for each part of 13.

12', 13' video signal υ, undercut 4,
It has been said that the magnetic particle pattern of the defect 3 mixed with the spatter 5 etc. is difficult to identify, and there is very little prospect of automatic discrimination.

The present invention has been proposed in view of the above circumstances, and its purpose is to be able to distinguish between defects and harmful pseudo-defects, reduce the area of visual inspection by hot-ray inspectors, and The purpose of the present invention is to provide a defect extraction method in MT inspection that facilitates automation.

The defect extraction method in MT inspection according to the present invention uses an ITV camera and an image processing device to perform absolute value difference processing of images before and after the magnetic particle inspection of the specimen, detects changes in the images of the specimen, and identifies defects. It is characterized by extracting magnetic particle patterns of harmful pseudo-defects, and pseudo-defects such as undercuts and spatters do not generate magnetic particle patterns due to excitation unless they contain harmful defects, but defects and harmful pseudo-defects Since magnetic particles remain with excitation, the shape of the defective part is clearly revealed, and the pseudo defective part is designed to automate visual inspection by taking advantage of the fact that concentration changes caused by residual magnetic particles. .

An embodiment of the present invention will be described in detail with reference to the accompanying drawings.

Figure 4 (B), (C), and Φ) are diagrams for explaining an embodiment of the method of the present invention, and Figure 4 (6) and 0) are diagrams for each specimen before MT inspection. , Figure 4 (Q, (
B) is a diagram showing the MT test results of each test piece, Figure 5
The figure shows one embodiment of the apparatus used to carry out the method of the present invention.

In Figure 4 (4), (B), (C) I (D), 14.11 is the specimen before MT inspection, 15.18 is the MT inspection result, 16 is a defect, and 19 is a harmful pseudo defect. 0 In FIG. 5, 9 is an ITV camera, 2o is an image processing device, 21 is a control signal, 22 is an image memory, 23 is an arithmetic unit, and 24 is a defect signal.

The operation of the above embodiment of the present invention will be explained.

In Figures 4 (A), (B), and (C), 14.77 is the specimen before the MT test, and specimen 14.
It is assumed that the specimen 17 has a defect, and the test specimen 17 has a harmful pseudo-defect. Letting these MT inspection results be is and is, respectively, in the case of 15, the magnetic particle pattern of defect 16 clearly appears. However, in the case of No. 18, although magnetic particles remain in the harmful pseudo defect 19, for example, in the undercut portion, there appears to be no significant change in appearance. However, even in this case, changes can be recognized if attention is paid to the fact that the magnetic particles have undercuts and differences in color density. Therefore, if IA-Cl=E or IB-Dl=E, a defect signal 24 indicated by E is obtained. This can be achieved by inputting the output signal of the ITV camera 9 to the image processing device 20 as shown in FIG. For details, see M.
Before and after the T test is controlled by a control signal 21, images of 14 or 11 test specimens for MT test are recorded on the image grid 22, and images 15 or 18 after the next MT test are recorded in the computing unit 23. A defect signal 24 can be obtained by comparing the values. The image processing device 20 shown in FIG. 5 can be of various types;
In short, it is sufficient to have a function that can perform absolute value difference processing of images before and after MT examination.

In view of the above, the method of the present invention provides the following excellent effects.

(1) By using the image absolute value difference method, which is an application of the image processing method, it becomes possible to automatically distinguish between defects and harmful pseudo-defects in the specimen during MT inspection.

(2) Automatic visual inspection becomes possible by introducing equipment for implementing the method of the present invention, that is, an ITV camera and an image processing device, into a conventional MT inspection line.

[Brief explanation of the drawing]

Figure 1), @, Figure 2 (G), 0) and Figure 3 (N
. (B) is a diagram for explaining the conventional example, and Figure 4 (
6), (B), (C), υ) are diagrams for explaining an embodiment of the method of the present invention, respectively, and FIG.
), @ is each specimen before MT inspection, Figure 4 (C'),
O)) is a diagram showing the MT test results of each test specimen, and FIG. 5 is a diagram showing an embodiment of the apparatus used to carry out the method of the present invention. 9... ITV camera, 14.17... Test specimen before MT inspection, 15.18... MT inspection result, 16o defect, 19... Harmful false defect, 20... Image processing device,
21... Control signal, 22... Image memory, 23...
- Arithmetic unit, 24... defective signal.

Claims (1)

[Claims] Performing absolute value difference processing of images before and after the magnetic particle flaw detection inspection of the test object using an ITV Kanut image processing device, detecting changes in the image of the test object, and extracting magnetic particle patterns of defects and harmful pseudo-defects. A defect extraction method in magnetic particle inspection, characterized by: