JPS6134447A - Optical flaw detector for continuously cast slab - Google Patents

Abstract

PURPOSE:To maintain stable, highly accurate flaw detecting capability all the time, by outputting the optimum shading-correction objective-level value and operation timing to a correction-factor operating circuit in correspondence with the change in shape of a picked up image signal. CONSTITUTION:The picked up image signal from a camera 4 is inputted to a signal smoothing and operating circuit 7 and a shading correcting circuit 10. In the signal smoothing and operating circuit 7, smoothing operation of the signal of each element of light receiving elements is carried out. The result of operation is outputted to a correction factor operating circuit 8. A shading correction control circuit 9 outputs an objective level and operation timing to the correcting-factor operating circuit 8 in correspondence with the shape of the picked up image signal. A constant is operated and the correcting constant is computed for every element. The signal from each corresponding element from the camera 4 is multiplied by the constant in the shading correcting circuit 10. The output of the correcting circuit 10 is inputted to a flaw judging circuit 11, where the flaw is judged.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an optical flaw detection device for detecting flaws in slabs in continuous casting equipment.

In conventional continuous casting equipment, in order to reliably guarantee slab surface quality, improve yield through completely maintenance-free rolling of slabs, and promote energy savings through thermal flaking, optical flaw detection equipment is used to detect slab surface conditions and perform flaw discrimination processing. It is being done.

Problems to be solved by this invention In order to demonstrate and maintain high-accuracy and stable flaw discrimination ability in such optical flaw detection equipment, it is necessary to use the COD, which is an imaging device.
It is important that the base level of an imaging signal from a camera or the like is constant in the slab width direction.

However, in reality, as shown in Figure 4(a), there are variations in the signal level in the width direction of the slab, and from such signals, conventional signal processing circuits are used to identify only the portions that are considered to be defects. When taken out, as shown in FIG. 4('b), there were cases where the flaw detection level could not be reached and the flaw detection was omitted.

As a means of solving such flaw detection omissions, there is a method of lowering the flaw determination level and determining even a small signal as a flaw, but this is not practical because the influence of signals from other parts that are not flaws is large. In contrast, the base signal from the COD camera, which is an imaging device, is corrected and adjusted to a certain level, and the signal is corrected so that a highly accurate and stable flaw inspection ratio can be performed in the subsequent flaw discrimination process. can be done.

As pre-processing for this correction process, it is necessary to perform a correction constant calculation process. In order to perform this correction constant calculation process, the operator focuses a COD camera, etc., and then obtains an image of the entire signal. Originally, it was necessary to manually perform calculation processing and adjust the camera focus again.

For example, when attempting to detect flaws of several centimeters or more across the entire width of a slab, the operator must calculate the correction constant for shading correction at intervals of 20 to 30 minutes, with an interruption time of approximately 5 minutes. I was taking it and doing it.

Therefore, not only the flaw detection is interrupted during this work, but also the work efficiency is poor, which is a practical problem.

Therefore, the present invention solves the above-mentioned technical problems in signal processing in conventional slab optical flaw detection, and enables correction constant calculation processing to be performed automatically and periodically in a signal processor without manual intervention by an operator. It is an object of the present invention to provide a slab optical flaw detection device that can always maintain stable and highly accurate flaw detection ability without interrupting flaw detection.

Means for Solving the Problems The slab optical flaw detection apparatus according to the present invention averages the signal level between predetermined sample points of the input imaging signal and outputs a smoothed signal when performing shape ink correction. Equipped with a smoothing calculation circuit, and can perform correction constant calculations. A correction coefficient is calculated in comparison with the i-ing correction target level value, and a shading correction circuit multiplies the image signal by the correction coefficient to perform shading correction. The present invention is characterized in that it is equipped with a shading correction control circuit that outputs an optimal shading correction target level value and calculation timing to a correction coefficient calculation circuit in accordance with the above.

FIG. 1 shows a control block diagram of this apparatus, and FIGS. 2 and 3 (a) and 3(a) and 3(b) illustrate waveforms at the can processing stage.

A light source 3 is installed at a predetermined position on the continuous casting line to emit light from diagonally above the slab 2 surface in the slab width direction. A CCD camera 4 is provided to take images.

This CCD camera 4 has a 2048-bit light receiving element, and the surface condition of the slab 2 is divided into 2048 element signals and input to the signal processing circuit IK.

The signal processing circuit 1 includes a signal smoothing calculation circuit 7. Correction coefficient calculation circuit 8. Shading correction circuit 10. It is composed of a shading correction control circuit 9 and a flaw discrimination circuit 11.

Among these, an image signal from the CCI camera 4 is input to a signal smoothing calculation circuit 7 and a shading correction circuit 10. As mentioned above, the CCD camera 4 has two light receiving elements.
Since a 048-bit silicon light receiving element is used, the signal from the CCD camera 4 is divided into 2048 element signals.

Here, in the signal smoothing calculation circuit 7, a smoothing calculation is performed on the signals of each element of the light receiving element in order to obtain an image of the entire imaging signal without losing focus of the CCD camera 4.

In this smoothing operation, as shown in FIG. 2, the signals of 11 adjacent elements of the imaging raw signal a are averaged, and the resulting value is taken as the representative value of the central element among the 11 elements. This averaging process is performed at intervals of 200 elements, and the overall image as shown in C is calculated by connecting the representative values of about 10 points obtained in this way.

The results of the smoothing calculation thus performed are output to the correction coefficient calculation circuit 8, and a total of 2048 correction constants are calculated for each element. For example, in FIG. 2, if the target level of shading correction is set as Cp, the level of the signal image for the nth element is Cn, so Kn=Cp/Cn (Kn>1), n'th element. For the th element, Km' = Cp /
Cn'(Kn' (1). - The constants from 1 to 2048 thus obtained are used as signals from the corresponding elements from the CCD camera 4 in the shading correction circuit 10 connected to the next stage. After being multiplied, a signal of a constant level as shown in FIG. 3(a) is generated.

The output of the shading correction circuit 10 is input to a flaw discrimination circuit 11 connected to the next stage. The flaw determination circuit 11 determines the presence or absence of flaws based on the flaw determination level d of the shading-corrected imaging signal inputted from the shading correction circuit 10, as shown in FIG. 3(b).

The output of the flaw discrimination circuit 11 is output in parallel to the printer 5 and the addition point 12, and the printer 5 prints out the flaw discrimination results. Furthermore, the imaging signal from the CCD camera 4 is directly input to the other input of the addition point 12.
The flaw determination results are superimposed on the slab 2 image and displayed on the monitor 6.

The signal processing section 1 also includes a shading correction control circuit 9.
is provided. This shading correction control circuit 9
outputs the target level Cp and calculation timing to the correction coefficient calculation circuit 8 for each slab molecule to perform constant calculation in accordance with the shape of the imaging signal that changes with changes in the casting situation and environment.

Furthermore, the signal processing circuit 1 is equipped with a switch SW, and this switch SWK transfers the image signal from the CCD camera 4 to the shading correction circuit 10 or the flaw discrimination circuit 1.
Enter either one of 1. .

When performing shading correction, switch SW is switched to the shading correction circuit 10 side as shown in FIG. It is also possible to switch to the flaw discrimination circuit 11 side and directly input a raw image signal without shading correction to the flaw discrimination circuit 11 to perform flaw discrimination. - In the above configuration, the imaging signal from the CCD camera 4 is smoothed by the signal smoothing calculation circuit 7 to calculate a signal C representing the entire image and output to the correction coefficient calculation circuit 8. The correction coefficient calculation circuit 8 calculates a constant of 1 to 2048 for each element based on the target level Cp inputted from the shading correction control circuit 9 and the calculation timing, and outputs it to the shading correction circuit 10.

The shading correction circuit 10 performs shading correction by multiplying constants corresponding to each element of the image pickup signal by 1 to 2048, and the flaw discrimination circuit 11 discriminates flaws based on the flaw determination level d.

The determination result of the flaw determination circuit 11 is printed out on the printer 5, and is superimposed with the imaging signal at the addition point 12 and displayed on the monitor 6.

Effects of the Invention The embodiment of the slab optical flaw detection device for continuous casting equipment according to the present invention is as described above, and can bring about the following effects.

By having a signal processor automatically and periodically calculate the correction constant for shading correction, not only is there no interruption in the operator's work and no interruption in the flaw detection of the flaw detection equipment, but there is also no need to worry about changes in the casting situation or environment. Therefore, stable and highly accurate flaw detection ability can always be maintained over the entire slab width direction.

[Brief explanation of the drawing]

Fig. 1 is a control block diagram showing an embodiment of the slab optical flaw detection device of the present invention, Fig. 2 is a waveform diagram showing an example of a signal image extracted from an imaging signal, and Fig. 3(a) is a waveform diagram after shading correction. , FIG. 3(b) is a waveform example diagram in the flaw discrimination circuit, and FIGS. 4(a) and (b) are waveform example diagrams according to the conventional method. l...Signal processing circuit, 2...Slab, 3...Light source, 4...COD camera, 5...Printer, 6...Monitor, 7...Signal smoothing calculation circuit, 8...Correction coefficient calculation circuit, 9
...Shading correction control circuit, 10...Shading correction circuit, 11...Flaw discrimination circuit, 12...Addition point, S,, S2...
Defect, a...Image signal, b...Representative value, C...Signal image, d...Flaw determination level, Cp...Target level, W...
Slab width. 2g (a) Figure 3 (a) (b) Figure 4 (b)

Claims (1)

[Scope of Claims] In an optical flaw detection device that images the surface of a slab with an imaging device and performs shading correction processing on the imaged signal and then performs flaw determination, the imaged signal is input and the signal level for each predetermined sample point section is determined. a signal smoothing calculation circuit that averages and outputs a smoothed signal; a correction coefficient calculation circuit that receives the smoothed signal from the signal smoothing calculation circuit and calculates a correction coefficient by comparing it with a shading correction target level value; and the correction coefficient. a shading correction circuit that performs shading correction on the image signal by inputting the correction coefficient from an arithmetic circuit; and a shading correction circuit that performs the correction by calculating an optimal shading correction target level value and calculation timing in response to a change in the shape of the image signal. A continuous casting slab optical flaw detection device characterized by being equipped with a shading correction control circuit that outputs to a coefficient calculation circuit.