JPS6191544A - Automatic exposure control for surface defect detection of hot metal material - Google Patents

Abstract

PURPOSE:To eliminate overlooking of defects, by controlling the light receiving time by varying the scanning cycle of a linear array image sensor to always keep the exposure sensitivity level of the linear array sensor constant. CONSTITUTION:An input signal converted into electricity from light from a linear array image sensor 1 for sensing heat radiation and a linear array image sensor 2 for sensing reflected light of an externally illuminated light on the surface of a cast billet is fed to an LPF7 to determine the extra polation line of the signal. From this extra polation line, readings during one scan are obtained by a pulse applied from a quantity of light data sampling control circuit 12 through a quantity of light sampling 8 and the data points averaged with an averaging circuit 9 to calculate the difference from the target quantity of light with a gain setting circuit 10. The alteration value of the necessary exposure time is calculated based on the deviation. The results of the calculation are inputted into a exposure time controlling circuit 11 to control the exposure time by changing the interval of the photoelectric conversion start pulse of linear array image sensors 1 and 2 and thus, the output signal of a flaw detection camera can be outputted with a fixed signal level.

Description

[Detailed Description of the Invention] (Industrial Application Field) In detecting surface defects in hot metal materials, thermal radiation emitted from the hot metal itself is imaged by a linear array image sensor of a flaw detection camera, and When the surface is irradiated with an external illuminator with a specific wavelength and this reflected light is subjected to Ia& by the linear array image sensor of the flaw detection camera, the light receiving sensitivity of the linear array image sensors is always maintained at a constant level, mutually or independently. In this specification, the results of the development of a procedure effective for significantly improving the adjustment range for light reception, that is, an exposure control method for detecting surface defects in hot metal materials, will be described hereinafter.

When the thermal radiation of a hot metal material, for example, a hot cast steel billet produced by continuous casting, is imaged using a linear array image sensor of a flaw detection camera, the heat radiation of the cast billet is uniform at all positions of the cast billet. This is required because if the thermal radiation is not uniform, the light receiving sensitivity of the linear array image sensor of the flaw detection camera will not be constant.

On the other hand, the surface of a hot metal material, such as a hot cast steel billet, is irradiated with an external illuminator having a specific wavelength, and this reflected light is detected by a linear array image sensor of a flaw detection camera.
When performing an RI image, the reflectance of the irradiated light changes depending on the surface properties of the cast steel piece.

(Prior art) Regarding the above points, according to the conventional method, the aperture of the imaging lens can be adjusted depending on the exposure sensitivity of the linear array image sensor of the flaw detection camera, but the adjustment range of the aperture adjustment is limited to the lens aperture of the flaw detection camera. Approximately 2.
The problem remains that the culmness is only 25 times (corresponding to the aperture value MAX 1.8), and the signal gain is corrected by the automatic gain control circuit (AGC) for the electrical signal photoelectrically converted by the exposure of the linear array image sensor. (The adjustment range by No. 8 IQ correction is about 15 times the conventional background noise component of a steel piece using a camera (signal level P-P-1.5V, back noise o, iv> and is still non-conforming.

The change in radiant light due to thermal radiation is that the spectral radiance of iron is approximately 80% when sampled with a wavelength of 1 μm as the standard, as shown in the following table.
The change is about 0 times (at 600 to 1200°C).

As described above, since the amount of change in thermal radiation is large, it is clearly impossible to keep the thermal radiation receiving sensitivity constant through adjustment using the lens aperture and self-disturbance.

(Problems to be Solved by the Invention) In general, the temperature distribution of a cast steel billet has a temperature gradient such as, for example, about 1000° C. at the center and 750° C. at the ends in the width direction of the cast steel billet.

Similarly, in the longitudinal direction of the cast steel piece, the temperature changes from time to time depending on the #R construction conditions.

The exposure sensitivity of the linear array image sensor of the 1rA camera that detects thermal radiation is unavoidably subject to wide sensitivity changes, especially because the thermal radiation increases or decreases depending on temperature changes in the longitudinal direction of the steel piece.

In other words, in areas of high temperature, the exposure sensitivity is saturated, while in areas of low temperature, exposure sensitivity is insufficient and cannot be output as a flaw detection signal.
All of the reasons why this point cannot be addressed by the prior art have been mentioned.

On the other hand, the factors that affect the surface reflectance of a cast steel billet are mainly the scales that occur on the surface of the cast billet.Of course, if there is a lot of scale, the reflectance will decrease, and therefore the linear array image sensor of the Detector 1n camera will Exposure sensitivity causes wide sensitivity changes due to fluctuations in reflectance.

That is, in areas where scale has not occurred or where scale has peeled off, the reflectance is high and therefore the exposure sensitivity is saturated, and in areas where most of the scale has occurred or where the scale layer is thick, the exposure sensitivity is insufficient and cannot be used as a flaw detection signal.

The growth of scale that occurs in cast steel slabs is not uniform due to various factors, and this tendency is particularly remarkable in the longitudinal direction of casting of cast steel slabs.

As mentioned above, the exposure sensitivity of the linear array sensor of the f/f search camera used in conditions where both the thermal radiation and the reflected light from the surface of the steel plate fluctuate is always unstable.
This was causing serious problems.

Therefore, we have taken measures to keep the exposure sensitivity constant for each or each of the linear array image sensors for detecting thermal radiation of flaw detection cameras and the linear array image sensors for detecting reflected light from external illumination, as described above. It is an object of the present invention to maintain and improve the integrity of flaw detection by excluding fluctuations in the sensitivity level due to fluctuation factors.

(Means for Solving the Problems) The present invention provides a method for detecting surface defects in hot metal materials using a flaw detection camera that simultaneously images radiant light emitted by the hot metal itself and reflected light from external illumination at the same position. , Linear array image sensor 2 used in flaw detection cameras
For exposure control 2U to equalize differences in imaging sensitivity, find the average level or li level of the sensitivity level of the amount of light received by the linear array image sensor that receives radiant light, and compare this with the radiant light target sensitivity level. , On the other hand, find the average sensitivity level of the amount of light received by the linear array image sensor that receives reflected light (or simply find the level and compare it with the target sensitivity level of reflected light, and calculate the correlation between radiant light sensitivity and anti-DJ light sensitivity. The light receiving time is controlled by changing the scanning period of the linear array image sensor based on the 1% deviation of the sensitivity or the deviation of the sensitivity level for each of the radiant light sensitivity and the reflected light sensitivity, and the exposure sensitivity of the linear array sensor is always maintained. This is an automatic exposure control method in a method for detecting surface defects in hot metal materials, which is characterized by keeping the level constant.

In order to maintain or improve the degree of flaw detection, it is necessary that the sensitivity level of the linear array image sensor is always at an appropriate level. As I have already made clear, there is no such thing.

Now, Fig. 1 shows the optical configuration of the flaw detection camera, in which 1 is a linear array image sensor for detecting thermal radiation, 2 is a linear array image sensor for detecting reflected light, 4 is a hot cast steel billet, 4 is a lens, 5 is a lens aperture adjustment device, and 6 is an image sensor drive amplifier circuit. When detecting a defect from the surface image of the cast steel billet 3 using the photo-electric conversion signals of the linear array image sensors 1 and 2, the defect signal S
/N becomes a critical extraction requirement.

The S/N of the defect signal is closely related to the sensitivity level of the linear array image sensor, and as shown in Figure 2, if the sensitivity level is not between 0.5 and 1 V at the background level, the S/N will be It gets exponentially worse.

The flaw detection camera consists of a linear array image sensor for detecting thermal radiation and a linear array image sensor for detecting reflected light from external illumination that includes wavelengths that do not interfere with thermal radiation, which are placed in the same optical system. Since the structure is such that layer images are formed at the same point, the exposure sensitivity must be controlled to maintain appropriate levels in response to various conditions of thermal radiation and reflected light from external illumination.

Furthermore, since the optical system must be the same for thermal radiation and reflected light, lens aperture control is limited due to differences in sensitivity levels.

Based on the relational structure described below, an exposure control method for keeping each sensitivity level constant is achieved by controlling the exposure period of each linear array image sensor 2II 71'.

The linear array image sensor 1 for sensing thermal radiation and the near array image sensor for sensing anti-Q and J light on the surface of a cast steel piece of external illumination light can be used together or individually, and the light amount formula for the amount of light received is the average light. Φ is determined, and this light m or the average '& target light quantity is compared, and the interval of photoelectric conversion start pulses of each linear array image sensor is changed based on the deviation thereof.

This allows the exposure time of the linear array image sensor to be controlled so that the sensitivity signal level remains constant.

More specifically, in FIG. 3, an extrapolation line of the photo-electrically converted input signal from the linear array image sensor is obtained using the UPF.

From this extrapolation line, the table value during one scan is applied through the light intensity sampling 8 with the applied pulse from the light intensity data sampling control circuit, and this data n points are averaged by the averaging circuit, and the layer with the target light intensity is calculated by the gain setting circuit. The amount of change in exposure time required is calculated based on the amount of deviation.

This calculation result is input to an exposure time control circuit to control the exposure time by changing the interval of photo-electrical conversion start pulses of the linear array image sensor.

In this way, the output signal of the flaw detection camera could always be output at a constant signal level.

(Function) In conventional surface defect detection equipment, the output signal of the flaw detection camera fluctuates due to the reasons mentioned above, so the S/N of the defective part is
However, the automatic exposure control of the present invention can sometimes cause defects to be overlooked if they pass through the field of view of the flaw detection camera when the S/N ratio is not sufficient due to fluctuations in the output signal. Since the output signal of the flaw detection camera is subjected to exposure time control based on deviation calculation and feedback, the output signal level is always constant and the S/N ratio can always be maintained in a good condition.

(Effects of the Invention) By incorporating automatic exposure control according to the present invention, the signal level can be kept constant, and defective parts are no longer overlooked, which conventionally occurred due to fluctuations in the signal level.

[Brief explanation of drawings]

Figure 1 is a skeleton diagram showing the optical configuration of the flaw detection camera, Figure 2 is a graph showing the relationship between the average plague level and S/N ratio of the linear array image sensor, and Figure 3 is a block diagram of the exposure control circuit. be. Figure 1 Figure 2 Figure 3

Claims (1)

[Scope of Claims] 1. In a method for detecting surface defects in hot metal materials using a flaw detection camera that simultaneously images radiant light emitted by the hot metal itself and reflected light from external illumination at the same position, the flaw detection camera is used. When performing exposure control to equalize the difference in imaging sensitivity between the two linear array image sensors that receive radiant light, find the average level or simple level of the sensitivity level of the amount of light received by the linear array image sensor that receives radiant light, and combine this with the radiant target sensitivity level. On the other hand, find the average level or simple level of the sensitivity level of the amount of light received by the linear array image sensor that receives reflected light, compare this with the reflected light target sensitivity level, and calculate the difference between the radiant light sensitivity and the reflected light sensitivity. The light reception time is controlled by changing the scanning period of the linear array image sensor based on the deviation of the sensitivity levels between each other or the sensitivity levels of the radiant light sensitivity and the reflected light sensitivity, so that the exposure sensitivity level of the linear array sensor is always maintained. 1. An automatic exposure control method in a method for detecting surface defects in hot metal materials, the method comprising: maintaining constant .