JP6021798B2 - Surface defect inspection equipment - Google Patents

Description

  The present invention relates to a surface defect inspection apparatus that inspects a surface defect generated on the surface of a rolled material, and more particularly to a surface defect inspection apparatus that inspects a surface defect of a long base material such as a wire, a steel bar, or a steel pipe.

  Conventionally, metal substrates such as wire rods, steel bars, and steel pipes are industrially manufactured by hot rolling. In this hot rolling, the substrate passing speed is often as high as 5 to 100 m / sec as compared with other processes, and visual defect inspection is often difficult. Therefore, measures such as automatically detecting surface defects using a non-destructive inspection method, removing parts where surface defects were detected, and correcting process conditions to suppress the occurrence of surface defects, etc. It is taken. As such a non-destructive inspection method, electromagnetic techniques such as an eddy current flaw detection method and a leakage magnetic flux flaw detection method are widely used.

  This eddy current flaw detection method detects the electromagnetic characteristics of the material to be inspected such as the permeability and conductivity of the base material continuously in the direction of material passing through, and surface defects occur when the detected electromagnetic characteristics change beyond the threshold. It is the structure which determines with having carried out. However, in the eddy current flaw detection method, noise increases when the substrate vibrates during inspection to cause lift-off fluctuations or when the electromagnetic characteristics are locally uneven on the surface of the substrate. In addition, there is a problem in that a larger number of surface defect signals are erroneously detected than actual defects, and the defect detection accuracy may be lowered.

Moreover, what is actually detected by the eddy current flaw detection method is an electrical signal such as amplitude and phase, and only information indicating that there is a surface defect of a certain size and depth. That is, in the eddy current flaw detection method, other information such as the shape of the surface defect is not detected, and it is difficult to identify what the detected surface defect is.
Further, the eddy current flaw detection method is configured to determine that a surface defect has occurred when there is a large change in electromagnetic characteristics that are continuously measured. Therefore, for example, when the change in electromagnetic characteristics is small, such as a defect having a length of a certain length or more in the material passing direction, no change in electromagnetic characteristics is observed at the intermediate part of the defect. In principle, it may be difficult to detect other than the front and rear ends.

  In order to compensate for the disadvantages of the eddy current flaw detection method as described above, an optical method using a high-speed camera, that is, an image flaw detection method may be used. By using such an optical technique, it is possible to obtain a merit that the surface defects can be classified from the detected image of the surface defects, but on the other hand, the pattern of the substrate surface and the unevenness of the optical reflectance are reduced. There is a possibility of erroneous detection as a surface defect, and there is a possibility that erroneous detection will occur in the same manner as in the eddy current flaw detection method.

Therefore, as shown in Patent Document 1 and Patent Document 2 below, a method combining an eddy current flaw detection method and an image flaw detection method is adopted.
For example, Patent Document 1 discloses a surface defect inspection technique that includes an image flaw detection unit on the downstream side of the eddy current flaw detection unit along the material passing direction, and a detection signal detected by the eddy current flaw detection device is used as a trigger. A technique for capturing an image of a part corresponding to a surface defect detected by an eddy current flaw detector by turning on the strobe illumination of the image flaw detector is disclosed. If both the electromagnetic characteristic permeability data and the image data are prepared for the surface defect thus generated, the occurrence of the surface defect can be determined based on the two detection results, and only the eddy current flaw detector is used. Compared to the case, it becomes possible to improve the detection accuracy of surface defects.

  In addition, in Patent Document 2, information on the defect depth obtained by the eddy current flaw detector is added to the information obtained by performing the continuous inspection using the image inspection device on the metal rod as the base material. A technique for improving the accuracy of defect determination by referring to both pieces of information is disclosed.

JP 2009-8659 A JP 2010-107513 A

However, neither the eddy current flaw detection method nor the image flaw detection method described above is perfect as a method for detecting surface defects.
For example, in the eddy current flaw detection method, noise generated due to various factors may be erroneously detected as a surface defect, and the reliability is not so high. In particular, since the magnetic permeability of the base material increases when the temperature of the base material falls below the Curie temperature, the hot-rolled material with temperature unevenness near the Curie point is likely to have uneven magnetic permeability, resulting in surface defects. Detection is also likely to occur. In addition, the eddy current flaw detection method also has a drawback in that it cannot sufficiently detect a surface defect having a length in the material passing direction of the substrate as described above.

On the other hand, the image flaw detection method does not have a detectability regarding the depth of the defect as compared with the eddy current flaw detection method. Therefore, if there is a pattern on the surface of the base material or if there is local unevenness in reflectance, a portion that is not defective may be erroneously detected as a defect.
That is, the two detection methods described above have low reliability with respect to the detected result, and it is unclear in Patent Document 1 and Patent Document 2 which of the two detection results should be adopted. ing. Although it is possible to investigate whether surface defects have occurred using the logical sum (OR) or logical product (AND) of the results detected by the two detection methods, it is possible to determine which data is reliable. Therefore, the occurrence of erroneous detection cannot be effectively reduced, and the detection capability of the inspection apparatus cannot be increased in many cases.

  The present invention has been made in view of the above-mentioned problems, and by using information other than the image obtained in the image flaw detection unit, masking erroneous detection occurring in the eddy current flaw detection unit, the detection accuracy of the surface defect An object of the present invention is to provide a surface defect inspection apparatus capable of improving the resistance.

In order to solve the above problems, the surface defect inspection apparatus of the present invention employs the following technical means.
That is, the surface defect inspection apparatus according to the present invention is a surface defect inspection apparatus that inspects a surface defect generated on the surface of a base material that has been hot-rolled in the material being passed, An eddy current flaw detection unit that continuously detects the electromagnetic characteristics of the substrate and detects the surface defect from the discontinuous change in the detected electromagnetic characteristics, and the eddy current flaw detection unit along the material passing direction of the substrate An image flaw detection unit that is arranged adjacent to the image sensor and picks up an image of the surface of the base material that is being passed through and detects the surface defect from the picked-up image, and a signal of the surface defect detected by the eddy current flaw detection unit On the other hand, a defect determination unit that performs defect determination while removing a false detection signal by performing correction based on information obtained by the image flaw detection unit is provided.

Preferably, the defect determination unit removes the false detection signal from the defect signal detected by the eddy current flaw detection unit using information on the vibration of the base material obtained by the image flaw detection unit. It is good to be taken.
Preferably, the defect determination unit removes the false detection signal from the defect signal detected by the eddy current flaw detection unit using information on the surface temperature of the base material obtained by the image flaw detection unit. It may be configured.

Preferably, the image flaw detection unit is configured to capture a self-luminous image of the base material.
The most preferable form of the surface defect inspection apparatus according to the present invention is a surface defect inspection apparatus that inspects a surface defect generated on the surface of the base material with respect to a hot-rolled base material during threading, An eddy current flaw detection unit that continuously detects the electromagnetic characteristics of the base material during threading, detects the surface defects from the discontinuous change in the detected electromagnetic characteristics, and the direction along the threading direction of the base material An image flaw detection unit that is arranged adjacent to the eddy current flaw detection unit, images the surface of the base material in the thread, and detects the surface defect from the imaged image, and the surface detected by the eddy current flaw detection unit A defect determination unit that performs defect determination while removing a false detection signal by performing correction based on information obtained by the image flaw detection unit with respect to a defect signal, and the defect determination Part relates to the vibration of the base material obtained in the image flaw detection part Using broadcast, from the signal of the defects detected in the eddy current testing unit, characterized in that it is configured to remove signals of the false detection.

  According to the surface defect inspection apparatus of the present invention, it is possible to improve detection accuracy of eddy current flaw detection by masking erroneous detection occurring in the eddy current flaw detection section using information other than the image obtained by the image flaw detection section. it can.

It is the figure which showed the surface defect inspection apparatus of 1st Embodiment. It is the figure which showed the eddy current flaw detection part. It is the figure which showed arrangement | positioning of the camera along the direction perpendicular | vertical to the material passing direction in an image flaw detection part. It is the figure which showed arrangement | positioning of the camera along the direction parallel to the material passing direction in an image flaw detection part. It is the figure which showed the positional relationship of the signal output from the eddy current test part, and the signal output from the image test part. It is the figure which showed the image of the front-end | tip part of the base material imaged by the image flaw detection part, and the image of a surface defect. It is the figure which showed the vibration state of the base material. It is the figure which showed the surface temperature of the base material.

[First Embodiment]
Hereinafter, embodiments of the surface defect inspection apparatus 1 of the present invention will be described in detail with reference to the drawings.
As shown in FIG. 1, the surface defect inspection apparatus 1 of 1st Embodiment is the surface defect which generate | occur | produces on the surface of this base material W with respect to the base material W which is hot-rolled by the hot rolling equipment and is moving. It is the structure which inspects.

Examples of the substrate W to be inspected by the surface defect inspection apparatus 1 include a long rolled material having a circular cross section such as a wire rod, a steel bar, a steel pipe, and the like, which are hot-worked. However, the base material W to be inspected may be a material such as an ingot or a rolled plate.
In the following, an inspection apparatus provided in a hot rolling facility for hot rolling a steel bar (hot steel) at a speed of 5 m / sec to 100 m / sec while heating to a temperature of about 900 ° C. to 1100 ° C. The surface defect inspection apparatus 1 of the present invention will be described by giving an example.

  Specifically, the surface defect inspection apparatus 1 described above continuously detects the electromagnetic characteristics of the substrate W (hot rolled material) in the thread, and detects surface defects from the transient changes in the detected electromagnetic characteristics. The eddy current flaw detection unit 2 that detects the surface of the substrate W is arranged adjacent to the eddy current flaw detection unit 2 along the direction of passing the base material W, and the surface of the base material W in the material passing is imaged. And an image flaw detector 3 for detecting a defect. In addition, the surface defect inspection apparatus 1 corrects the surface defect signal detected by the eddy current flaw detection unit 2 based on the information obtained by the image flaw detection unit 3, thereby removing a false detection signal. It has the defect determination part 4 which performs a defect determination.

Next, the eddy current flaw detector 2 and the image flaw detector 3 constituting the surface defect inspection apparatus 1 will be described.
As shown in FIG. 2, the eddy current flaw detection unit 2 is a penetration type, and continuously measures the electromagnetic characteristics of the passing substrate W, and when there is a predetermined change in the measured electromagnetic characteristics, It is configured to determine that a surface defect has occurred. Specifically, the eddy current flaw detection unit 2 converts an vortex current generated by the eddy current generated in the coil unit 7 when the base material W passes through the inside of the coil unit 7 to be described later into a DC electric signal and outputs the eddy current. It has a flaw detection circuit 5 and eddy current defect detection means 6 for judging the presence or the nature of surface defects based on the electrical signal sent from the eddy current flaw detection circuit 5.

  The eddy current flaw detection circuit 5 is a bridge circuit including two through-type coil portions 7 formed by winding one conductive wire in the opposite winding direction by the same number of turns. The eddy current flaw detection circuit 5 is configured such that the base material W continuously penetrates the winding centers of the two coil portions 7, and when the base material W passes through the winding centers of the respective coil portions 7, As a result, an induced electromotive force is generated in the coil section 7. Therefore, if it can be confirmed that a current flows through the coil portion 7, it can be determined that a surface defect has occurred.

  Specifically, since the winding direction of the conducting wire is reversed in these two coil parts 7, the induced electromotive force generated in one coil part 7 and the other coil part 7 are generated in a normal base material part. The induced electromotive force is different from each other only in the direction of current with the same magnitude. Therefore, when there is no flaw and a substantially homogeneous base material W is passed, the bridge circuit of the eddy current flaw detection circuit 5 is balanced and no potential difference occurs. However, when the substrate W having a surface defect such as a scratch passes through any one of the coil portions 7, a difference occurs in the induced electromotive force between the eddy current flaw detection circuit 5, in other words, the two coil portions 7. A potential difference is generated between the two. The potential difference generated between the two coil portions 7 increases toward one side when the substrate W having a surface defect passes through one coil portion 7, and the substrate W passes through the other coil portion 7. When doing, it becomes larger in the opposite direction.

Therefore, the eddy current flaw detection circuit 5 is provided with a vector voltmeter 8 (also called a lock-in amplifier) that measures a voltage generated between the two coil portions 7. The change in voltage measured by the vector voltmeter 8 of the eddy current flaw detection circuit 5 is sent to the eddy current defect detecting means 6 as an electric signal.
The eddy current defect detecting means 6 determines the presence or absence of surface defects and the properties based on the electrical signal output from the vector voltmeter 8 of the eddy current flaw detection circuit 5 (voltage generated at the bridge between the two coil portions 7 described above). doing. That is, when the substrate W having a surface defect does not pass through the eddy current flaw detection circuit 5, the voltage output from the vector voltmeter 8 becomes zero, but the substrate W having a defect has the eddy current flaw detection circuit 5. When passing, a voltage is generated between the two coil portions 7 due to the difference in induced electromotive force. Since the amplitude of the voltage at this time becomes large when the size of the surface defect is large, the size of the defect can be determined from the magnitude of the amplitude. The amplitude and phase of the voltage detected by the eddy current defect detecting means 6 in this way are sent to the defect determining unit 4 as an electric signal (eddy current flaw detection signal described later).

  As shown in FIGS. 3 and 4, the image flaw detector 3 captures an image of the surface of the substrate W using a camera 9 (self-luminous image acquisition means), and generates surface defects based on the captured image. Is detected. The image flaw detection unit 3 is arranged on the upstream side or the downstream side of the eddy current flaw detection unit 2 along the material passing direction of the base material W, and detects a surface defect of the base material W that moves in the same manner as the eddy current flaw detection unit 2. It can be done. Specifically, the image flaw detection unit 3 is preferably disposed close to the eddy current flaw detection unit 2 so that the distance L is 2 m or less, and a substrate such as a roll is provided between the two. It is preferable that a member for restraining W is not provided. If the image flaw detector 3 and the eddy current flaw detector 2 are brought close to 2 m or less in this way, the correspondence between the surface defect data detected by the eddy current flaw detector 2 and the surface defect data detected by the image flaw detector 3. As a result, the ability to detect surface defects at the image inspection section 3 can be enhanced.

Further, the image flaw detection unit 3 includes a plurality of cameras 9 provided around the moving base material W and defect image detection means 10 that detects defects from images captured by these cameras 9. .
The cameras 9 described above are provided at equal intervals in the circumferential direction along the outer peripheral surface of the base material W so that the outer peripheral surface of the base material W can be imaged without omission. 3 and 4, four cameras 9 are arranged at equal intervals around the base material W, but the number of cameras 9 may be three or five or more. Also good.

  The camera 9 used in the image flaw detection unit 3 includes an optical filter that can capture a self-luminous image (self-luminous red-hot image), and a line camera that continuously photographs every line is used. That is, the base material W such as a hot-rolled rolled material has a higher temperature than a cold-rolled material or the like, and radiates infrared light (self-luminous) by itself. Therefore, if an image of the base material W that emits light by these cameras 9 is captured, it is not necessary to separately prepare a light source for capturing the image, and the configuration of the apparatus can be made compact. .

  The defect image detection means 10 combines the images of the surfaces picked up by the four cameras 9 into one image data, performs image analysis on the combined images, and extracts only images containing surface defects. It is the composition to do. In this defect image detecting means 10, first, a captured image is binarized based on a predetermined threshold. In this way, since there is a contrast difference with respect to the part where there is generally no surface defect, it is possible to separate the part considered to be a surface defect and the part considered not to have a surface defect on the screen. . When the area of the portion considered to be a surface defect obtained in this way satisfies a predetermined size, it is determined that the surface defect exists in the captured image. The “image including the surface defect” extracted in this way is also sent to the defect determination unit 4.

By the way, the surface defect inspection apparatus 1 of the present invention performs a correction based on the information obtained by the image flaw detection unit 3 on the signal of the surface defect detected by the eddy current flaw detection unit 2, thereby generating a false detection signal. A defect determining unit 4 that performs defect determination while removing the defect is provided. Specifically, the surface defect inspection apparatus 1 is picked up by the vibration detecting means 11 for detecting the vibration of the substrate W from the image picked up by the camera 9 of the image flaw detector 3 and the camera 9 of the image flaw detector 3. Temperature detecting means 12 for detecting the temperature of the substrate W from the obtained image. And the defect determination part 4 uses the information regarding the vibration of the base material W detected by the vibration detection means 11, and / or the information regarding the temperature of the base material W detected by the temperature detection means 12, and uses the eddy current flaw detection part 2 The false detection signal is removed from the defect signal detected in (1).

Next, the vibration detection unit 11, the temperature detection unit 12, and the defect determination unit 4 that are features of the present invention will be described.
The vibration detection means 11 detects the vibration of the base material W by detecting how much the imaging position of the base material W shown in the image captured by the camera 9 vibrates on the image. It is. Specifically, the vibration detection means 11 specifies the position of the edge of the base material W in the image using the above-described binarization processing or the like, and the base material W is continuously captured as the image is continuously taken. It is detected over time how the position of the edge of the surface changes. The change with time of the edge of the substrate W detected in this way is sent to the defect determination unit 4 as information related to “vibration of the substrate W” detected by the position vibration detection means 11.

The temperature detection unit 12 is configured to detect the temperature of the base material W by detecting the luminance of the base material W in the image captured by the camera 9. The reason for using such temperature detecting means 12 is as follows.
In other words, as a cause of erroneous detection in eddy current flaw detection, there is an uneven magnetic permeability due to a temperature drop of the substrate W. In the case of a magnetic base material W such as a steel material, the magnetism of the base material W is lost and the permeability becomes 1 at a temperature equal to or higher than the Curie point (about 780 ° C. in the case of steel material). If the temperature is further reduced, the permeability tends to increase rapidly. For this reason, when the substrate W having temperature unevenness near the Curie point is inspected by eddy current flaw detection, the eddy current flaw detection signal also fluctuates greatly due to large magnetic permeability unevenness. That is, there are many false detections when eddy current flaw detection is performed on the substrate W in the temperature range near the Curie point. Therefore, the temperature of the substrate W is measured by the temperature detection means 12 described above.

  Specifically, the temperature detection unit 12 is configured to detect the temperature from the average luminance of the base material W on the captured image. That is, based on the same principle as that of the radiation thermometer, in the image captured by the camera 9 described above, the self-luminous luminance is higher as the temperature of the object is higher, and lower as the temperature is lower. Therefore, the temperature detection unit 12 measures the variation of the average luminance in the width direction of the full length image, and when the variation of the average luminance exceeds a preset threshold value, an electrical signal indicating that a false detection may have occurred. Is output. The electrical signal detected by the temperature detection unit 12 is also sent to the defect determination unit 4 in the same manner as the electrical signal output from the vibration detection unit 11.

  As shown in FIG. 1, the defect determination unit 4 includes an electric signal output from the eddy current flaw detection unit 2, an electric signal output from the image flaw detection unit 3, an electric signal output from the vibration detection unit 11, and a temperature detection unit 12. The final surface defect is determined on the basis of the four electrical signals output from. Of the four input signals input to the defect determination unit 4, the input signals input from the eddy current flaw detection unit 2 and the image flaw detection unit 3 relate to surface defects, but the vibration detection unit 11 and the temperature detection unit 12. The input signal input from is used for removing a false detection signal.

  Specifically, in the defect determination unit 4, information on the vibration of the substrate W obtained by the image flaw detection unit 3, in other words, “positional change of the edge of the substrate W with time” detected by the vibration detection unit 11. By using the information regarding, the signal of the false detection caused by the vibration of the substrate W is removed from the signal of the defect detected by the eddy current flaw detection unit 2. That is, even if a signal indicating that a surface defect is present in the eddy current flaw detection unit 2 is obtained, if “vibration of the substrate W” is confirmed by the vibration detection unit 11 described above, this signal is deleted. False detection is reduced.

  Further, the defect determination unit 4 uses the information on the temperature of the substrate W obtained by the image flaw detection unit 3, in other words, the information on the “temperature of the substrate W” detected by the temperature detection unit 12, and uses the eddy current flaw detection unit 2. The signal of the false detection resulting from the temperature of the base material W is removed from the signal of the defect detected in (1). That is, even when a signal indicating that a surface defect is present in the eddy current flaw detection unit 2 is obtained, it is confirmed that the “temperature of the base material W” is lowered to a predetermined temperature or less by the temperature detection unit 12 described above. In some cases, erroneous detection is reduced by deleting this signal.

The above-described error detection signal removal based on the vibration of the substrate W and the error detection signal removal based on the temperature of the substrate W may be performed either alone or both.
Next, the surface defect detection method performed by the surface defect inspection apparatus 1 of the present invention, in other words, the surface defect inspection method of the present invention will be described.
As shown in FIG. 1, when a substrate W having a surface defect is passed through the surface defect inspection apparatus 1, first, in the eddy current flaw detection circuit 5 of the eddy current flaw detection unit 2, a difference in induced electromotive voltage between the two coil parts 7 occurs. It is output to the eddy current defect detecting means 6 by a bridge circuit and a vector voltmeter. The eddy current defect detection means 6 performs filtering for extracting only an appropriate frequency component after phase adjustment from the amplitude and phase signal of the input voltage, and outputs the result to the defect determination unit 4 as an eddy current flaw detection signal.

If this eddy current flaw detection signal is shown as the vertical axis and the horizontal axis as the passing time axis from the top of the substrate W, it can be shown as a change with time as shown in FIG. 5 can be viewed as a position in the longitudinal direction of the base material W because the base material W has a constant speed of about 10 m / sec.
In the example shown in FIG. 5, a plurality of peak-like voltage changes are observed at the tip portion up to the vicinity of the passing time of 6 seconds. Any of these peak voltage changes can be considered to be highly likely to be a surface defect.

  Here, it can be considered that the peak height of the voltage change indicates the size of the surface defect. In this example, the eddy current flaw detection signal due to the artificial defect having a diameter of 2 mm and a depth of 0.2 mm provided on the base material is adjusted in advance so as to be 100%. That is, when the relative voltage (ECT level) with φ2 mm defects as 100% is shown, the eddy current flaw detection signal with a relative voltage up to 25% is a small surface defect (S size), and the relative voltage is 25% to 50%. Eddy current flaw detection signals are standard surface defects (M size), eddy current flaw detection signals with a relative voltage of 50% to 100% are large surface defects (L size), and eddy currents with a relative voltage of 100% to 250%. The flaw detection signal can be classified into four levels by the surface defect having the largest size (LL size).

  The substrate W that has passed through the eddy current flaw detection unit 2 is sent to the image flaw detection unit 3 located on the downstream side of the eddy current flaw detection unit 2, and an image is picked up by the image flaw detection unit 3. That is, images (self-luminous images) captured by the four cameras 9 provided in the image flaw detection unit 3 are sent to the defect image detection means 10, and these four images are combined in the defect image detection means 10. An image obtained by imaging the outer peripheral surface of the material W over the entire circumference is obtained. Next, image processing such as binarization processing is performed on this image to extract surface defects. An electrical signal indicating the presence of the extracted surface defect (hereinafter referred to as an image flaw detection signal) is output to the defect determination unit 4. When this image flaw detection signal is shown superimposed on the graph of FIG. 5, the image flaw detection signal is confirmed at a position indicated by “*” in the figure, and the eddy current flaw detection signal from the eddy current flaw detection unit 2 and the image flaw detection unit 3 are confirmed. It can be seen that the image flaw detection signal from No.1 matches well.

  The image actually picked up by the image flaw detection unit 3 is as shown on the upper side of FIG. 6, and the base material W is shown by changing the contrast according to the temperature. That is, a high-temperature part in the image is red-hot and is imaged with high brightness, so that it is displayed brightly on the image. On the other hand, as shown in the lower side of FIG. 6, when there is a surface defect such as a fold, only the place with the surface defect is locally cooled and imaged black at a low temperature. The occurrence of a defect can be recognized. Note that black images may be captured locally due to other factors on the image, and it is not possible to judge all low-luminance parts on the image as defective, but the image has a length of about 10 mm or more. In addition, it can be determined that the place where the luminance changes rapidly is likely to be a surface defect.

By the way, in some cases, the surface defect is not actually confirmed by the image flaw detection unit 3 in the portion that is a surface defect in the eddy current flaw detection unit 2.
For example, a large eddy current flaw detection signal is obtained at the position indicated by “A” in FIG. 5 (the position at the top of the base material W and the passing time is about 0.1 sec), but the image is defective. No surface defects could be found from the actual steel. It is considered that the erroneous detection of the eddy current flaw detection signal is caused as follows.

That is, in the eddy current flaw detection unit 2 having the two coil parts 7 described above (the eddy current flaw detection unit 2 of the penetration type differential coil system), when the base material W vibrates rapidly, the base material W that penetrates the eddy current flaw detection circuit 5 is used. Inclination occurs. If it does so, a difference will arise between the two coil parts 7 in the distance of the coil part 7 and the base material W which penetrates this coil part 7, and a big eddy current test signal will be generated also in a part without a surface defect.

  Actually, in the eddy current flaw detection signal indicated by “A” in FIG. 5, the leading portion of the base material W vibrates greatly, and it is highly possible that the eddy current flaw detection portion 2 is erroneously detected due to the vibration of the base material W. It is judged. When such a false detection occurs, the threshold for detecting the eddy current flaw detection signal must be set high, and the eddy current flaw detection signal due to vibration must be masked, resulting in a decrease in detection accuracy. No longer.

Therefore, in the surface defect inspection apparatus of the present invention, the false detection generated in the eddy current flaw detection signal from the eddy current flaw detection unit 2 is removed using the signal relating to the vibration of the substrate W detected by the vibration detection means 11 described above. Yes.
Specifically, an edge (edge) is detected from a full length image of the base material W input to the defect image detection means 10 using a simple image processing method such as binarization processing, and the time for the material passing time is detected. The change in the position of the edge in the image is the vibration curve of the substrate W. If the vibration curve of the base material W detected in this way is applied with a frequency filter (bandpass filter) equivalent to that of the eddy current flaw detector, and if a large vibration remains even after filtering, vibration is generated. The eddy current flaw detection signal generated at the remaining time is determined to be erroneous detection.

  If the eddy current flaw detection signal is masked using the vibration of the base material W in this way, it is possible to extract only surface defects while reducing false detection due to the vibration of the base material W included in the eddy current flaw detection signal. Thus, the detection accuracy of the eddy current flaw detection unit 2 can be improved. For example, FIG. 7 shows the result of an eddy current flaw detection signal masking erroneous detection caused by vibration of the substrate W. Also in FIG. 7, a large vibration is observed when the passing time is around 1.9 seconds, and it is determined that an erroneous detection due to the vibration occurs in the eddy current flaw detection signal at this position.

  In general, when the vibration is severely generated in only a small part such as the head part or the tail part in the longitudinal direction of the base material W, it is often masked and used as a dead zone. However, according to the present invention, it is possible to reduce false detection without changing the threshold value when extracting surface defects from the eddy current flaw detection signal, and without providing a dead zone regardless of the presence or absence of vibration. Loss of accuracy can be prevented.

Further, it is desirable to apply not only image inspection and eddy current flaw detection but also visual inspection to a portion where masking is performed for the eddy current flaw detection signal.
On the other hand, as a false detection factor in eddy current flaw detection, the above-described uneven magnetic permeability due to a local temperature drop of the substrate W can be cited. Therefore, in the surface defect inspection apparatus 1 of the present invention, the erroneous detection generated in the eddy current flaw detection signal from the eddy current flaw detection unit 2 is removed using the temperature of the base material W detected by the temperature detecting means 12 described above. .

  Specifically, as shown in FIG. 8, the average luminance is calculated by averaging the screen luminance over the entire surface of the image of the substrate W input to the defect image detecting means 10, and calculating the average luminance. Is measured over time. For example, paying attention to the position where the passing time in FIG. 8 is about 12 seconds, it is determined that only the average luminance of the screen at this position is reduced and the temperature of the substrate W is reduced. Therefore, if an eddy current flaw detection signal indicating a surface defect is detected at the same position where the temperature decrease is confirmed, it is determined that the eddy current flaw detection signal indicates false detection. Therefore, if the eddy current flaw detection signal is masked in the defect determination unit 4, it is possible to reduce erroneous detection of the eddy current flaw detection unit 2.

According to the surface defect inspection apparatus 1 described above, the misdetection occurring in the eddy current flaw detection unit 2 is masked using information other than the image obtained by the image flaw detection unit 3, that is, information on the vibration and temperature of the substrate W. As a result, the detection accuracy of eddy current flaw detection can be improved.
Specifically, if the surface defect inspection apparatus 1 described above is used, a reliable surface defect that extracts and reduces false detection factors in eddy current flaw detection and complements the defects of eddy current flaw detection in the hot rolling process. The inspection apparatus 1 can be provided. Therefore, as a result, mass disposal of products due to erroneous detection of surface defects can be suppressed, which can contribute to yield improvement.

If the detection of surface defects is performed by combining the results from the image flaw detector 3 after reducing false detection of the eddy current flaw detector 2 as described above, the types of surface defects and the like can be classified.
The arrangement order of the eddy current flaw detection unit 2 and the image flaw detection unit 3 described above can be arbitrarily changed as long as the eddy current flaw detection unit 2 and the image flaw detection unit 3 are installed in close proximity so that vibration and temperature conditions can be determined to be substantially the same.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. In particular, in the embodiment disclosed this time, matters that are not explicitly disclosed, for example, operating conditions and operating conditions, various parameters, dimensions, weights, volumes, and the like of a component deviate from a range that a person skilled in the art normally performs. Instead, values that can be easily assumed by those skilled in the art are employed.

DESCRIPTION OF SYMBOLS 1 Surface defect inspection apparatus 2 Eddy current flaw detection part 3 Image flaw detection part 4 Defect determination part 5 Eddy current flaw detection circuit 6 Eddy current flaw detection means 7 Coil part 8 Vector voltmeter 9 Camera 10 Defect image detection means 11 Vibration detection means 12 Temperature detection means W group Material

Claims (3)

A surface defect inspection apparatus for inspecting a surface defect generated on the surface of the base material against a hot-rolled base material in the passing material,
An eddy current flaw detection unit that continuously detects the electromagnetic characteristics of the base material in the threading material and detects the surface defect from a discontinuous change in the detected electromagnetic characteristics;
An image flaw detection unit that is arranged so as to be adjacent to the eddy current flaw detection unit along the material passing direction of the base material, images the surface of the base material in the material passing, and detects the surface defect from the captured image; ,
A defect determination unit that performs defect determination while removing a false detection signal by performing correction based on information obtained by the image flaw detection unit on the surface defect signal detected by the eddy current flaw detection unit; ,
A has,
The defect determination unit is configured to remove the false detection signal from the defect signal detected by the eddy current flaw detection unit using information on the vibration of the base material obtained by the image flaw detection unit. Characteristic surface defect inspection device. The defect determination unit is configured to remove the false detection signal from the defect signal detected in the eddy current flaw detection unit using information on the surface temperature of the base material obtained in the image flaw detection unit. The surface defect inspection apparatus according to claim 1 . The surface defect inspection apparatus according to claim 2 , wherein the image inspection unit is configured to capture a self-luminous image of the base material.