JPS63184047A - Flaw detection apparatus - Google Patents

Abstract

PURPOSE:To enhance detection accuracy, by accurately confirming the position of an object to be inspected in order to inspect the appearance of the object to be inspected during movement and setting an accurate inspection region. CONSTITUTION:The image of a moving object to be inspected is photographed as a stationary image by a TV camera 2 and a stroboscope 3 and the image data digitized by an A/D converter part 9 is divided into two directions and one of them is stored in a frame memory 4 as the image corresponding to one picture and the other is inputted to a processing circuit 5 through a change- over device 6. Differentiation and thinning processings are performed in synchro nous relation to the clock signal of the camera 2 in the circuit 5 to send the thinning data to the first memory 7. After one picture is processed, the change- over device 6 is changed over from the memory 4 to the image and image data is again sent to the circuit in synchronous relation to the clock signal of the camera 2. Differentiation and thinning processings are performed in the circuit 5 in such a state that a threshold value is changed to send the thin ning data to the second memory 8. By performing the confirmation processing of a flaw part by CPU 13 using the thinning data obtained in the memories 7, 8, the detection reliability and accuracy of a flaw can be enhanced.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to a defect detection device for visually inspecting a moving object to be inspected.

(Background Art) Conventionally, in order to inspect the appearance of an object to be inspected, an image of the object to be inspected is photographed using a television camera, the image signal is differentiated, and the line is thinned at a predetermined level to perform defect inspection. A method has been proposed. In this conventional example, there is a large difference in the differential value between the edge part and the defective part of the object to be inspected.
When matching the thinning level for extracting defective parts, there is too much thinning information for recognizing the edges of the object to be inspected, making it difficult to accurately recognize the edges and resulting in misjudgments. The problem is that there are many cases where

(Object of the invention) The present invention has been made in view of the above points, and
The purpose is to provide a defect detection device that accurately recognizes the position of the inspected object and increases detection accuracy by setting the correct inspection area in order to visually inspect the moving inspected object. There is something to do.

(Disclosure of the Invention) In the defect detection device according to the present invention, FIGS.
As shown in the figure, there is a television camera 2 and a strobe 3 for inputting a still image of a moving inspection object 1, a frame memory 4 for storing the input image, and a differentiation and thinning of the input image in real time. A processing circuit 5, a switch 6 for switching and inputting the image signal from the television camera 2 and the image signal from the frame memory 4 to the processing circuit 5, and a switch 6 for inputting the image signal input from the television camera 2 to the processing circuit 5 to be inspected. A first memory 7 stores the result of line thinning at a level L1 that allows the edge portion 11 of the object 1 to be recognized, and a defective portion 12 of the object 1 that receives the image signal input from the frame memory 4 to the processing circuit 5.
and a second memory 8 for storing the result of thinning at a level L2 at which the line can be extracted.

Since the present invention is configured in this manner, the outline of the object to be inspected is recognized using the data thinned in the first time, the position of the object to be inspected and the area to be inspected for defects are determined, and the line thinned in the second time. It becomes possible to search for data including the defective portion extracted in the previously obtained area only, and it is possible to efficiently extract defects in a moving inspected object.

FIG. 1 is a block diagram showing a schematic configuration of a defect detection device according to an embodiment of the present invention. TV camera 2 is the 30th
As shown in FIG. 1, a moving inspection object 1 is photographed. A/D
The converter 9 converts an analog image signal obtained by the television camera 2 into a digital signal. In this case, the number of bits of the digital signal is arbitrarily set depending on the object. The frame memory 4 stores the image data converted into the digital signal described above. The switch 6 switches between an image signal directly input from the television camera 1 and an image signal read out after being stored in the frame memory 4 and sends the same to the processing circuit 5. The processing circuit 5 performs differentiation and line thinning processing at a speed synchronized with the clock signal (vertical synchronization signal) of the television camera 1.

The strobe 3 is turned on only momentarily to photograph the moving object 1 as a still image. The controller 10 controls the emission timing of the strobe 3 and the switching of images input to the processing circuit 5.

The first memory 7 is a memory for storing a thinned image obtained by extracting the edge portion 11 of the inspection target ¥@1. The second memory 8 is a memory for storing a thinned image including the defective portion 12 of the object 1 to be inspected.

The CPU 13 recognizes the edge portion 11 of the object 1 to be inspected from the thinning data and sets an inspection area, recognizes the defective portion 12, and determines the quality of the object 1 to be inspected.

The operation of this embodiment will be explained below. As shown in FIG. 3, an image of the inspected object 1 moving in the direction of arrow P is photographed as a still image by a combination of a television camera 2 and a strobe 3, and the image signal is A/D converted. . The image data digitized by the A/D converter 9 is divided into two directions; one is stored in the frame memory 4 as an image for one screen, and the other is input to the processing circuit 5 via the switch 6. The processing circuit 5 performs differentiation and thinning processing in synchronization with the clock signal of the television camera 2, and sends thinning data to the first memory 7.

FIG. 4(a) is a diagram showing an example of thinning data obtained in the first memory 7. In the figure, 11 is the object to be inspected 1
This is the edge part.

FIG. 5 is an explanatory diagram of the operation of differentiation and thinning processing. The waveform in the figure shows the differential value e of the image signal with respect to the line AA' of the object to be inspected 1 shown in FIG. Ll is a threshold value for edge extraction, and L2 is a threshold value for defect extraction.

After the processing for one screen is completed, the switch 6 is immediately switched to the image from the frame memory 4, and the image data is sent to the processing circuit 5 via the switch 6 again in synchronization with the clock signal of the television camera 2. .

The processing circuit 5 performs differentiation and thinning processing again by changing the threshold value, and sends the thinning data to the second memory 8. FIG. 4(b) is a diagram showing an example of thinning data obtained in the second memory 8. In the figure, 12 is a defective portion of the object 1 to be inspected.

Using the thinning data obtained in the first memory 7 and the second memory 8, the CPU 13 performs recognition processing for the defective portion 12.

FIG. 2 is a diagram showing the above processing timing.

Figure 2(a) shows the vertical synchronization signal of the television camera 2;
b) is the image processing start signal, (e) is the light emission signal of strobe 3, (d) is the image memory signal, (e) is the memory image output signal, and (1 (f) is the differential line). (g) is the timing of the thinning process using the threshold value L1, (h) is the timing of the thinning process using the threshold value L2, and (i> is the timing of the thinning process using the threshold value L1).
3 shows the timing of image processing according to 3.

(Effects of the Invention) As described above, according to the present invention, an image of a moving object to be inspected is captured as a still image using a strobe and a television camera, is stored in a frame memory, and is input from the television camera to a processing circuit. At the same time, the image signal input from the frame memory to the processing circuit is thinned to a level that allows the detection of defective parts of the object to be inspected. Recognize the outline of the object to be inspected using the previously thinned data, determine the position of the object and the area to be inspected for defects, and first obtain data including the defective areas extracted in the second thinning. It becomes possible to search only within the area, which has the effect of improving the reliability and accuracy of defect detection. In addition, since the first differentiation and thinning process was performed directly using the image signal captured from the television camera,
High-speed processing is possible, and the frame memory is 1
It has the advantage that it only needs to cover the screen.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of a defect detection device according to an embodiment of the present invention, FIG. 2 is an explanatory diagram of the same operation, and FIG.
The figure is a perspective view showing the schematic configuration of the optical system used in the above.
Figures (a) and (b) are diagrams respectively showing images obtained in the first memory and second memory used in the same, and FIG. 5 is an explanatory diagram of the operation of the processing circuit used in the same. 1 is an object to be inspected, 2 is a television camera, 3 is a strobe, 4 is a frame memory, 5 is a processing circuit, 6 is a switch, 7 is a first memory, and 8 is a second memory.

Claims (1)

[Claims] (1) A television camera and strobe for inputting still images of a moving object to be inspected, a frame memory for storing input images, a processing circuit for differentiating and thinning input images in real time, and a television camera. A switch that switches and inputs the image signal from the frame memory and the image signal from the frame memory to the processing circuit, and a switch that thins the image signal input from the television camera to the processing circuit to a level that allows the edges of the object to be inspected to be recognized. It is characterized by comprising a first memory for storing the image signal and a second memory for storing the result of thinning the image signal inputted from the frame memory to the processing circuit to a level that allows extraction of defective parts of the object to be inspected. Defect detection equipment.