JPH0238955A - Surface flaw inspection device - Google Patents

Abstract

PURPOSE:To decide the kind of a flaw automatically by processing a digital image which is inputted to an image processing part through the image pickup operation of a camera. CONSTITUTION:A sensor part 5 detects the position of the flaw previously and a control part 6 determines image pickup timing according to its detection information and makes a stroboscopic device 2 emit light to reset the synchronous signal to the camera. Further, a driving part 7 is moved to stop the camera 3 and stroboscopic device 2 at image pickup positions. An image processing part 9 outputs the synchronous signal to the camera 3, inputs and stores the video signal from the camera its internal image memories, and processes its images to decide the kind of the flaw, thereby displaying the result on a display monitor 10.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention images a material to be inspected such as a steel material moving at high speed,
The present invention relates to a surface flaw inspection device that inputs a still image and performs an external appearance inspection using the image.

[Conventional technology]

FIG. 4 shows a schematic configuration of an example of a conventional surface flaw inspection device as shown in, for example, Japanese Patent Application Laid-Open No. 54-1182811.
shows. In Figure 4, (1) is the moving specimen, (2
) is a strobe, and (3) is an industrial television camera.

@ is the field of view. At this time, the field of view is set to a fixed value, so that the width of the specimen is also increased. c13
is a sensor that detects the moving speed of the test material. (to) is a synchronous circuit, and the strobe (
The light emission of 2) and the imaging of the camera (3) are controlled so as to be performed in synchronization. The video signal that is the output of the camera (3) enters the signal processing circuit ().
In the signal processing step 1c), flaws and edges are emphasized. The video signal output from the signal processing circuit @ is sent to the scan converter c! IJK is input. @ displays one screen worth of images stored in the monitor and scan/'':1/verter@.

The inspector looks at this image and performs a visual inspection. The conventional surface flaw inspection apparatus taken as an example here performs an external appearance inspection by displaying an image of the surface of a moving object as a still image.

[Problems to be Solved by the Invention] First, in the conventional apparatus as shown in FIG. 4, the presence or absence of flaws and the presence of flaws are mainly determined visually. However, the camera (3
) and displays the image on a monitor, the visual appearance varies depending on the size of the flaw, the contrast, etc. When an inspector visually inspects such a monitor display screen, it is necessary to carefully watch the entire screen so as not to overlook anything, which causes considerable fatigue for the inspector. Further, the dispersion of the judgment criteria also increases. In order to solve these problems, the present invention automatically determines the presence or absence of a flaw and the type of flaw by performing image processing on an input image, thereby improving inspection accuracy.

[Means to solve the problem]

The surface flaw inspection apparatus according to the present invention includes an image processing section that performs image processing such as binarization on image data input from a camera and automatically determines a button based on the shape of a flaw.

[Effect]

According to this invention, by performing image processing such as binarization on a digital image taken by a camera and inputted into an image processing section, the type of flaw in the next screen after being taken is automatically determined. It is determined that

〔Example〕

Examples of the present invention will be described below.

FIG. 1 is a diagram showing an example of a surface flaw inspection apparatus according to the present invention. In the figure (11 is a moving specimen.

(2) is a strobe, (3) is a high-definition camera, and (4) is an imaging field of view. (5) The sensor section detects the position of the flaw in advance by a method other than optical image processing. (6) is a control unit that determines the imaging timing based on the information from the sensor unit (5), outputs a light emission command to the strobe (2), and in synchronization with this, sends a synchronization signal to the camera to the image processing unit. Apply glue cent. Furthermore, the drive unit (7) is moved to stop the camera and strobe at the imaging position. (7) is a drive unit that is equipped with a strobe (2) and a camera (3), moves on a trajectory (8) based on signals from a control unit (6), and stops at a position specified by a sensor unit (5). . (9) is an image processing unit, which first outputs a synchronization signal to the camera (3).

A video signal from a camera is input, A/D converted and stored in an internal image memory, and the type of flaw is determined by performing image processing on this image. M7'jQ[l is a result display monitor and displays the result of determining the type of flaw. Furthermore, αυ is a high-definition display that displays the flaw image captured by the camera (3).

Next, the operation will be explained.

First, a flaw is detected in advance using a laser or the like in a cell section (5) installed upstream of the imaging point of the camera on the conveyance path of the material to be inspected, and a detection signal is output together with its position information. The sensor section (5) constantly detects the conveyance speed at the same time as detecting flaws. The control unit (6) inputs the flaw detection signal and conveyance speed information from the sensor unit (5), and first outputs the K drive command to the drive unit (7) to detect flaws in the strobe (2) and camera (3). Before passing the imaging point, the width direction of the material to be inspected (direction perpendicular to the conveyance path) is set in advance and the material is moved to the ninth position.On the other hand, the control unit (6) makes sure that the detected flaw is exactly in front of the camera. Tracking is performed until the flaw passes in front of the camera, and at the moment the flaw passes in front of the camera, a strobe (2) K light emission command is output, and strobe (2)
) to emit light. At the same time, an external synchronization reset signal is output to the image processing section (9). The camera (3) constantly captures images using an external synchronization signal from the image processing unit (9), and the first image processing unit (9) adjusts the timing of this capture, that is, the rising edge υ of the external synchronization signal, from the strobe. In order to match the light emission timing, the external synchronization signal is reset by a reset signal from the control section (6). camera(
In step 3), imaging is performed based on this synchronization signal, and the second image processing section (9) takes in the image and determines the type of flaw. This flaw type determination result is displayed on a result output monitor. Furthermore, since the next image taken by the camera is displayed on a high-definition display, it is possible to confirm the image in which the type of defect has been determined.

FIG. 2 shows the image processing section (9) in detail. In diagram M2, α2 is a synchronization circuit that generates a synchronization signal based on commands from the control unit (6), and α3 is a camera (
3) a camera interface that outputs a synchronization signal and inputs an image signal; 64 is an image memory;

α5 is a preprocessing unit that removes noise from the image stored in the image memo IJ a41, a binarization unit that binarizes the image data of CL (mcL4), and αD is a preprocessing unit that removes noise from the image stored in the image memo IJ a41. 0g is a result output unit that outputs the flaw type determination result to the result output monitor α0.

The image processing unit (9) is configured in this way. First, the synchronization circuit α2 resets the synchronization signal based on the reset signal from the control unit (6), and matches the flash timing of the strobe with the timing of the synchronization signal of the camera. let

This synchronization signal is outputted to the high-definition camera (3) through the camera interface α3, and an image signal corresponding to the synchronization signal is inputted, and this image signal is FiA/D converted and stored in the image memory α4. The preprocessing unit α9 performs preprocessing such as noise removal on the data in the image memory, and the preprocessed image data is then binarized by the binarization unit (Ill).Flaw type determination unit In αη, the next image is binarized and the linear 1
The type of flaw, such as dotted or one-sided flaw, is determined and displayed on the result output monitor through the result output section α seconds.

Next, other embodiments of the invention will be described.

FIG. 3 is a diagram showing another embodiment of the invention.

In this embodiment, the configuration of the device is exactly the same as in the above example, but the processing contents of the image processing device (9) are different. In contrast, in this embodiment, the reference image is first subtracted from the input multivalued image.

The obtained image is binarized and defects are determined. In the figure (31, f61. +13 to α9 are the same as in Fig. 2. a9 is a reference image, which is an image of a non-defective part of the target material to be tested. You can set an image or
Furthermore, it is also possible to set the image immediately before the sensor unit determines that there is a defect during online execution as the non-defective part. ω
The Fi subtractor subtracts the data of the reference image 12α9 from the data of the image memo 1JQ41 as shown in the following equation to create an image for flaw determination.

Xl,j = Ui,j -R1,j In this formula or actually, the density of the image to be determined by binarization, U is the density of the image input from the camera, R
is the density of the β black image. Also, the subscript 1 and j are the first on the screen.
Indicates that it is a pixel in row and column j. Qll is a switch that switches between displaying the input image from the camera and displaying the next image output from the X-reducer.

According to the present embodiment, even if a shadow that is easily confused with a flaw appears regularly on the image due to the shape of the material to be inspected, it can be removed and the flaw discrimination accuracy can be improved.

〔Effect of the invention〕

As described above, according to the present invention, the burden on inspectors is reduced by providing the image processing unit with an automatic flaw type discrimination function.
Stable inspection accuracy can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram of an embodiment of this invention, FIG. 2 is an explanatory diagram showing a detailed configuration example of an image processing section, and FIG. 3 is a detailed configuration example of an image processing section of another embodiment of this invention. FIG. 4 is an explanatory diagram showing the conventional embodiment! This is a diagram N. In the figure, (1
) is the material to be inspected, (2) is the strobe, (3) is the high-definition camera, (41i is the imaging field of view according to the present invention, (5) is the sensor section, (6) is the control section, and (7) is the drive section. (8) is the orbit, (9) is the image processing unit, (1 (l is the monitor for displaying the results, +111 is the high-definition display, U
is a synchronous circuit. 0 is the camera interface, -α aneurysm is the image memory. a5 is a preprocessing unit, aa is a binarization unit, and αη is a defect determination unit. 111 is a result output unit, (L9 is a reference image, ■ is a subtractor,
+211 is a switch. Note that the same symbols in Figure 9 indicate corresponding parts.

Claims (2)

[Claims] (1) A surface flaw inspection device that captures an image of a moving object, inputs it as a still image, and uses that image to perform an external inspection.A strobe that illuminates the material to be inspected and a high-definition device that inputs an image of the surface of the material to be inspected. a driving unit that is equipped with the above-mentioned strobe and high-definition camera and can move over the material to be inspected, a sensor unit that detects the imaging position on the line, and a sensor unit that detects the position information obtained by this sensor unit. originally outputs a drive command to the drive unit, and simultaneously calculates the imaging timing,
A control unit that outputs a strobe light emission command to the strobe and further calculates the timing of a reset command for an imaging synchronization signal, and a control unit that outputs a synchronization signal to the camera and captures an image signal from the camera as a still image and detects defects through image processing. An image processing unit that performs the determination of defects, and a CRT that displays the results of the determination of defects.
A surface flaw inspection device comprising a monitor and a high-definition display that displays images of flaws processed by the image processing section. (2) In the surface flaw inspection apparatus according to claim (1), there is provided a reference image memory for storing an image of a non-defective part as a reference image, and a memory for determining a flaw after extracting a defective part. A surface flaw inspection device comprising: a subtracter that subtracts the preset reference image from a captured image.