JP2013057570A - Surface flaw inspection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect flaws on the surface of an inspected object even when the surface condition of a harmless pattern is unstable while preventing erroneous detection of the harmless pattern.SOLUTION: A surface flaw inspection device 1 includes an imaging unit 4 that receives reflectance of illumination light applied to an inspected object 2 and takes a surface image of the inspected object 2, a polarization angle adjustment unit 12 that changes relative angles of a polarizer 5 and an analyzer 6 for defining optical condition of the imaging unit 4, an image monitor 10 that evaluates the surface image of the inspected object 2 taken by the imaging unit 4, and a polarization angle correction input unit 11 that inputs angle of the analyzer 6 according to evaluation by the image monitor 10.

Description

  The present invention relates to a surface defect detection apparatus that optically detects defects existing on the surface of an object to be inspected, such as a thin steel plate surface.

  Conventionally, a surface that optically detects defects existing on the surface of the object to be inspected by irradiating the surface of the object to be inspected, such as a thin steel plate surface, and analyzing the reflected light from the surface of the object to be inspected. Defect detection devices are known. Such an optical surface defect detection apparatus may erroneously detect a harmless pattern that is similar to a defect existing on the surface of the object to be inspected but is not a detection target as a defect. For this reason, a technique for suppressing erroneous detection of this harmless pattern has been proposed.

  For example, in Patent Document 1, light is incident on the surface of an object to be inspected at an incident angle of 35 degrees to 75 degrees, and an image of a first camera installed in the regular reflection direction of reflected light, the incident direction or the normal direction. A technique is described that suppresses false detection of harmless patterns by comparing with an image of a second camera installed in an angle direction within 20 degrees from the reflection direction. In Patent Document 2, a polarizing plate and a quarter-wave plate are formed by configuring an optical system that optimally adjusts the polarization state of light reflected from the object to be inspected using a polarizing plate and a quarter-wave plate. A technique is described that suppresses false detection of harmless patterns by adjusting.

JP 58-204353 A JP 2005-221391 A

  However, the technique described in Patent Document 1 is effective when the state of the harmless pattern on the surface of the object to be inspected is constant, but the state of the harmless pattern is changed, and the two cameras are both harmless. In the case of detection, there is a problem that a harmless pattern is erroneously detected. Moreover, since the technique described in Patent Document 2 designs an optical system so as to cancel the harmless pattern in advance, there is a problem in that it is not possible to suppress the false detection of the harmless pattern when the state of the harmless pattern is not constant. there were.

  The present invention has been made in view of the above, and even when the state of the harmless pattern on the surface of the object to be inspected is not constant, it is possible to detect a defect on the surface of the object to be inspected without erroneously detecting the harmless pattern. An object of the present invention is to provide a surface defect detection apparatus.

  In order to solve the above-described problems and achieve the object, a surface defect detection device according to the present invention receives reflected light of illumination light irradiated on an object to be inspected and captures a surface image of the object to be inspected. Imaging means for performing, changing means for changing an optical condition for imaging by the imaging means, evaluation means for evaluating the surface image of the object to be inspected imaged by the imaging means, and the changing means according to the evaluation by the evaluation means And adjusting means for controlling.

  According to the surface defect detection device of the present invention, it is possible to detect a defect on the surface of the inspection object without erroneously detecting the harmless pattern even when the state of the harmless pattern of the inspection object is not constant.

FIG. 1 is a schematic diagram showing the configuration of the surface defect detection apparatus according to the first embodiment of the present invention. FIG. 2 is a schematic diagram showing an example of the angle adjustment mechanism of the analyzer. FIG. 3 is a diagram for explaining the polarization state of reflected light when linearly polarized light is irradiated on the steel sheet surface having a harmless pattern. FIG. 4 is a diagram for explaining the reason why the harmless pattern looks different depending on the angle of the analyzer. FIG. 5 is a diagram for explaining the timing for adjusting the angle of the analyzer in the inspection of the pickling line. FIG. 6 is a schematic diagram showing the configuration of the surface defect detection apparatus according to the second embodiment of the present invention. FIG. 7 is a diagram for explaining an example of an analyzer angle automatic setting method using the noise level as an index. FIG. 8 is a graph in which the noise level is graphed for each angle of the analyzer.

  Hereinafter, an embodiment of a surface defect detection device according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

[First Embodiment]
FIG. 1 is a schematic diagram showing the configuration of the surface defect detection apparatus according to the first embodiment of the present invention. As shown in FIG. 1, a surface defect detection apparatus 1 according to an embodiment of the present invention includes an illumination unit 3 that irradiates an inspection object 2 with illumination light, and a reflection from the inspection object 2 irradiated with the illumination light. An imaging unit 4 that receives light, and a polarizer (polarizing plate) 5 and an analyzer (polarizing plate) 6 that define optical conditions for imaging by the imaging unit 4 are provided.

  The illuminating unit 3 includes an optical system in which a light beam is guided through an optical fiber from a light source 7 using, for example, a metal hydrolamp, and the illuminating unit 3 irradiates the inspection object 2 with the light beam. A polarizer 5 having a polarization angle set to 45 degrees is disposed in the optical path of the irradiation light between the emission end of the illumination unit 3 and the inspection object 2. As a result, linearly polarized light including polarization components in the vertical direction (P-polarized light) and the horizontal direction (S-polarized light) is irradiated onto the device under test 2.

  The incident angle of the illumination light emitted from the illumination unit 3 to the object to be inspected 2 can be set to 60 degrees, for example. On the other hand, the imaging unit 4 is arranged at a position for receiving regular reflection light of irradiation light irradiated from the illumination unit 3 to the inspection object 2. That is, the reflection angle of the reflected light received by the imaging unit 4 is also 60 degrees. The incident angle and the reflection angle are not limited to 60 degrees, and may be other angles as long as the difference between the P-polarized light and the S-polarized light of the reflected light is easy to detect.

  An analyzer 6 is disposed in the optical path of the reflected light between the object to be inspected 2 and the imaging unit 4, and the imaging unit 4 receives the reflected light through the analyzer 6. Note that the angle of the analyzer 6 is provisionally set in advance according to the manufacturing conditions of the inspection object 2 to an angle that can suppress the erroneous detection of the harmless pattern.

  By the way, even if the angle of the analyzer 6 is set according to the manufacturing conditions of the object 2 as described above, the harmless pattern actually changes due to variations in the manufacturing conditions. Misdetection may occur. Therefore, in the surface defect detection apparatus according to the first embodiment of the present invention, the operator manually changes the angle of the analyzer 6. Hereinafter, a configuration for that purpose will be described.

  As shown in FIG. 1, the surface defect detection device according to the first embodiment of the present invention includes an image input unit 8 that controls an image signal output from the imaging unit 4, and an image that is controlled by the image input unit 8. An image processing unit 9 that performs image processing on the signal, and an image monitor 10 that displays the image signal processed by the image processing unit 9 as an image are provided. Furthermore, the surface defect detection apparatus according to the first embodiment of the present invention includes a polarization angle correction input unit 11 for an operator to set the angle of the analyzer 6, and the analyzer 6 according to the input from the polarization angle correction input unit 11. And a polarization angle adjusting unit 12 for driving the angle.

  The image input unit 8 includes an amplifier that amplifies the image signal output from the imaging unit 4, a filter that cuts electrical noise, and an A / D converter that converts the image signal that is an analog signal into an image signal of a digital signal. Are provided inside. The image processing unit 9 has a preprocessing function for performing shading correction processing and noise removal processing using a spatial filter in order to remove background effects such as uneven reflectance of the background from the input image signal. The image monitor 10 is a general image display device such as a CRT or LCD.

  In addition to the above configuration, the surface defect detection apparatus according to the first embodiment of the present invention calculates a feature amount such as a length, width, and area of a defect candidate from the output image of the image processing unit 9. 13, a defect determination unit 14 that determines the type and degree of the defect from the feature amount by a predetermined algorithm, an external input / output unit 15 that displays or outputs the type and degree of the determined defect, and an external input / output unit 15 Is provided with a host computer 16 for performing quality control of the inspected object 2 by using the output from. In addition, the surface defect detection apparatus according to the first embodiment of the present invention includes a light amount control unit 17 that controls the output of the light source 7 in accordance with the image input unit 8.

  FIG. 2 is a schematic diagram illustrating an example of an angle adjustment mechanism of the analyzer 6. For example, if a rotation mechanism having a hole in a ring shape is provided at the incident end of the imaging unit 4 and the analyzer 6 is attached to the rotation mechanism, the angle of the analyzer 6 is arbitrarily set to 360 degrees. Can be set to an angle.

  Hereinafter, a method for adjusting the analyzer 6 in the surface defect detection apparatus according to the first embodiment having the above-described configuration will be described.

  On the surface of the pickled steel sheet, a harmless pattern having the same shape as the lashes to be detected, such as scale residue or red scale, is generated. The pickled steel sheet refers to a steel sheet that has been subjected to a pickling process in which a steel sheet is passed through a liquid tank containing a strong acid such as hydrochloric acid or sulfuric acid to remove scale on the surface. What remains without being completely removed in this pickling treatment is the remaining scale. Although there is an application for detecting the remaining scale itself, in the following, as an inspection for detecting defects on the surface of the object 2 to be inspected, an inspection for detecting lashes on the surface of the steel plate on the pickling line is assumed. Is described as harmless.

  FIG. 3 is a diagram for explaining the polarization state of reflected light when linearly polarized light is irradiated on the steel sheet surface having this harmless pattern. As shown in FIG. 3 (a), the surface defect detection device according to the first embodiment of the first embodiment of the present invention irradiates a steel plate as an object to be inspected with linearly polarized light having a polarization angle of 45 degrees. . In FIG. 3A, the polarization state of the linearly polarized light irradiated on the steel plate is indicated by a straight line with a symbol A.

  FIG.3 (b) is a figure which shows the polarization state of the reflected light of the linearly polarized light irradiated to the steel plate. As shown in FIG. 3B, the polarization state of the reflected light differs depending on whether the linearly polarized light is reflected by the base of the steel plate or the linearly polarized light is reflected by the remaining scale on the steel plate. In FIG. 3 (b), the polarization state of the elliptically polarized light reflected by the base of the steel plate is indicated by an ellipse indicated by symbol B, and the polarization state of the elliptically polarized light reflected by the harmless pattern is indicated by an ellipse indicated by symbol C. The reason for the difference in the polarization state of the reflected light is that the linearly polarized light is irradiated depending on whether the linearly polarized light is reflected by the base of the steel plate or the linearly polarized light is reflected by the remaining scale on the steel plate. This is because the action on the S polarization component is different from that on the S polarization component.

  The surface defect detection apparatus 1 according to the first embodiment of the first embodiment of the present invention, using the above properties, adjusts the angle of the analyzer 6 disposed in the optical path of the reflected light, thereby preventing harmless pattern errors. Suppress detection. FIG. 4 is a diagram for explaining the reason why the harmless pattern looks different depending on the angle of the analyzer 6.

  As shown in FIG. 4A, when the angle of the analyzer 6 is not appropriate (the angle of the alternate long and short dash line in the figure), the elliptically polarized light reflected by the base of the steel plate and the elliptically polarized light reflected by the rest of the scale on the steel plate Then, the magnitude | sizes of the polarization component which permeate | transmits the analyzer 6 differ. This is because the size of the component that passes through the analyzer 6 becomes a projected component of elliptically polarized light with respect to the angle of the analyzer. In FIG. 4 (a), the direction of polarized light passing through the analyzer 6 is indicated by a chain line, and the projected component of elliptically polarized light reflected by the base of the steel plate is indicated by a dashed double-pointed arrow. The projected component of the reflected elliptically polarized light is indicated by a solid double arrow. In the example shown in FIG. 4 (a), it can be seen that the broken double arrow and the solid double arrow are different in size, and the base of the steel plate and the harmless pattern are contrasted and imaged.

  On the other hand, as shown in FIG. 4B, when the angle of the analyzer is appropriate (solid line of angle α in the figure), the elliptically polarized light reflected by the base of the steel plate and the elliptically polarized light reflected by the remainder of the scale on the steel plate. And the sizes of the components that pass through the analyzer are equal (or substantially equal). Similar to FIG. 4 (a), in FIG. 4 (b), the projected component of the elliptically polarized light reflected by the base of the steel plate is indicated by a dashed double-pointed arrow, and the projected elliptically polarized light reflected by the remaining scale on the steel plate. Ingredients are indicated by solid double arrows. In the example shown in FIG. 4B, it can be seen that the broken line double arrow and the solid line double arrow are approximately equal in size, and the base of the steel sheet and the harmless pattern are imaged without any contrast.

  For the above reasons, the surface defect detection apparatus 1 according to the first embodiment of the present invention can suppress harmless pattern false detection by adjusting the angle of the analyzer 6 as shown in FIG. it can. However, the angle of the analyzer 6 as shown in FIG. 4B varies depending on the manufacturing conditions and the remaining scale. Therefore, when the angle of the analyzer 6 is fixed, a situation occurs in which erroneous detection of a harmless pattern cannot be suppressed. Therefore, in the surface defect detection apparatus 1 according to the first embodiment of the present invention, the operator confirms the occurrence state of the harmless pattern on the image monitor 10 at the timing when the manufacturing conditions of the continuously manufactured steel sheet change, and the analyzer. The angle of the analyzer 6 is adjusted by inputting the angle 6 to the polarization angle correction input unit.

  Hereinafter, the timing at which the operator should adjust the angle of the analyzer 6 will be described using an example of detection of defects on the surface of the steel sheet on the pickling line.

  FIG. 5 is a diagram for explaining the timing for adjusting the angle of the analyzer 6 by detecting defects on the surface of the steel sheet on the pickling line. Fig.5 (a) is a schematic diagram on the pickling line provided with the surface defect detection apparatus 1 which detects the defect of the steel plate surface. As shown in FIG. 5A, in the pickling line, the steel plate that is the object to be inspected 2 is carried into the pickling line while being wound as a coil 18. Then, by rewinding the coil 18, the steel sheet is passed through the pickling line and subjected to the pickling treatment, and then the surface defect detection device 1 detects a defect on the surface of the steel sheet. The object to be inspected 2 for which detection of this defect has been completed is wound up by a coiler to become a winding coil 19 and is carried out of the pickling line, and the pickling line is completed.

  Since the pickling line shown in FIG. 5A is a continuous line, the steel plate as the object to be inspected 2 is welded and continuously passed through the pickling line. Therefore, as shown in FIG. 5 (b), the coil A and the coil B of the steel plate as the object 2 to be inspected are detected by the surface defect detection device 1 in a state of being integrated through the welded portion 2b. Is implemented.

  From the above, when the surface defect detection device 1 according to the first embodiment of the present invention is used for detection of defects on the surface of a steel plate on a pickling line, the welded portion 2b that welds between the coils of the steel plate as the inspection object 2. It can be seen that the operator should reset the angle of the analyzer 6 immediately after passing through the surface defect detection device 1.

  Specifically, the operator confirms the surface image of the harmless pattern occurrence status confirmation area 2a in the vicinity of the downstream side of the welded portion 2b on the image monitor 10, operates the polarization angle correction input device 11, and operates the harmless pattern occurrence status. The angle of the analyzer 6 is adjusted so that the harmless pattern in the confirmation area 2a is minimized. In this way, by adjusting the angle of the analyzer 6 at the timing when the welded portion that welds the coils passes through the surface defect detection device, the harmless pattern can be erroneously detected even if the occurrence of the harmless pattern changes. Therefore, it is possible to inspect the surface of the object 2 to be inspected.

  In the above description, the embodiment of the present invention has been described using the embodiment in which the optical condition is changed by changing the angle of the analyzer 6. However, for example, the positions of the illumination unit 3 and the imaging unit 4 are determined. Even if the optical conditions are changed by changing the incident angle and the reflection angle of the illumination light by driving, the present invention can be appropriately implemented without changing the gist of the present invention.

  As described above, the surface defect detection apparatus according to the first embodiment of the present invention receives the reflected light of the illumination light irradiated on the inspection object 2 and images the surface image of the inspection object 2. The polarization angle adjusting unit 12 that changes the relative angle between the polarizer 5 and the analyzer 6 that defines the optical conditions for the imaging unit 4 to capture an image, and the surface image of the object 2 to be inspected captured by the imaging unit 4 are evaluated. In accordance with the evaluation by the image monitor 10 and the evaluation by the image monitor 10, the operator includes a polarization angle correction input unit 11 for inputting the angle of the analyzer 6, and the deflection angle adjusting unit 12 is set to the angle of the input analyzer 6. Since the angle of the analyzer 6 is controlled, it is possible to inspect defects on the surface of the object to be inspected without erroneously detecting the harmless pattern even if the harmless pattern occurrence state changes.

[Second Embodiment]
FIG. 6 is a schematic diagram showing the configuration of the surface defect detection apparatus according to the second embodiment of the present invention. In the first embodiment of the present invention, a method in which an operator manually finds the optimum angle of the analyzer is used. However, in the second embodiment of the present invention, for example, the noise level of a steel plate is used as an index, and this noise level is minimized. This is an embodiment in which the angle is automatically set to the optimum angle of the analyzer.

  As shown in FIG. 6, the surface defect detection apparatus according to the second embodiment of the present invention has substantially the same configuration as the surface defect detection apparatus according to the first embodiment of the present invention. Therefore, in the description of the surface defect detection apparatus according to the second embodiment of the present invention, the same components as those of the surface defect detection apparatus according to the first embodiment of the present invention are denoted by the same reference numerals, and the description of the same components is omitted. Shall.

  The surface defect detection apparatus 1 according to the second embodiment of the present invention includes an image evaluation unit 20 in addition to the components of the surface defect detection apparatus 1 according to the first embodiment of the present invention. The image evaluation unit 20 is an arithmetic unit that calculates a noise level in the image of the object 2 to be inspected, which is an output from the image processing unit 9, and is realized by a general MPU. Therefore, in FIG. 6, the image evaluation unit 20 is separated from the image processing unit 9, but the image evaluation unit 20 and the image processing unit 9 may be configured integrally.

  Here, as the noise level of the image, for example, the standard deviation σ of the luminance of each pixel can be used. Hereinafter, an image noise level using the standard deviation σ of luminance of each pixel is abbreviated as a noise level σ, and description will be made using the noise level σ.

  In addition, the image evaluation unit 20 transmits a control signal to the polarization angle adjustment unit 12 according to the calculated noise level of the image. As the control of the polarization angle adjustment unit 12 performed by the image evaluation unit 20, for example, the following method can be considered.

  7 and 8 are diagrams for explaining an example of an automatic method for setting the angle of the analyzer 6 using the noise level σ as an index.

  In the example of the method for automatically setting the angle of the analyzer 6 according to the second embodiment of the present invention, the surface defect detection device 1 displays the image of the harmless pattern occurrence state confirmation area 2a (see FIG. 5B) as the analyzer 6. The angle is acquired while changing the angle at a pitch of 0 to 10 degrees. That is, in the example of the method for automatically setting the angle of the analyzer 6 according to the second embodiment of the present invention, as shown in FIG. 7, the harmless pattern occurrence status confirmation corresponding to each angle of 10 degrees pitch is confirmed. An image of area 2a is acquired. In the example shown in FIG. 7, it is assumed that the angle of the analyzer 6 at the timing when the weld 2b passes the surface defect detection device 1 is preliminarily set to 0 degrees as an angle that suppresses false detection of harmless patterns. The harmless pattern occurrence status confirmation area 2a is illustrated. Further, a certain length of the harmless pattern occurrence status confirmation area 2a corresponding to the 10 degree pitch of the analyzer 6 is defined as an L pitch. It should be noted that how many ranges the angle of the analyzer 6 is examined before and after a preset angle may be determined by a parameter in order to perform the processing in a short time (that is, a short distance on the steel plate). good.

  The image evaluation unit 20 calculates the noise level σ of the image of the harmless pattern occurrence state confirmation area 2a obtained at each L pitch as described above, and uses the angle 6 as the noise level σ of the analyzer 6 corresponding to each L pitch. Remember. As the image for calculating the noise level σ, either an uncorrected image or an image normalized by performing shading correction in the image processing unit 9 may be used.

  FIG. 8 is a graph in which the noise level σ calculated as described above is graphed with respect to each angle of the analyzer 6.

  In the example of the graph shown in FIG. 8, the noise level σ is minimum when the angle of the analyzer 6 is around 70 degrees. Therefore, in the example of the graph shown in FIG. 8, it can be seen that the harmless pattern contrast is minimized when the angle of the analyzer 6 is 70 degrees. Therefore, in this case, the image evaluation unit 20 may control the polarization angle adjustment unit 12 so that the angle of the analyzer 6 is 70 degrees.

  Note that the search algorithm for the angle of the analyzer 6 according to the second embodiment of the present invention can use a general search algorithm such as a hill-climbing method. That is, the noise level σ at the previous L pitch is compared with the noise level σ at the current L pitch, and if the noise level σ decreases, the noise level σ at the next L pitch is calculated. It is possible to use an algorithm that searches for the L pitch at which the noise level σ starts to increase by performing the process of repeating the above. In the example of the graph shown in FIG. 8, the noise level σ decreases to the nth L pitch, and the noise level σ increases from the (n + 1) th L pitch. This reveals that the noise level σ of the nth L pitch is the minimum value. Of course, it is possible to further divide the space between the nth L pitch and the (n + 1) th L pitch to improve the accuracy.

  As described above, the surface defect detection apparatus according to the second embodiment of the present invention receives the reflected light of the illumination light irradiated on the inspection object 2 and captures the surface image of the inspection object 2. The noise level of the surface image of the inspected object 2 is calculated by calculating the polarization angle adjusting unit 12 that changes the relative angle between the polarizer 5 and the analyzer 6 that define optical conditions for the imaging unit 4 to perform imaging. And an image evaluation unit 20 that searches for the angle of the analyzer 6 that minimizes the noise level as a control angle. The polarization angle adjustment unit 12 controls the analyzer 6 so that the angle of the analyzer 6 becomes the searched control angle. Since the angle is controlled, it is possible to detect a defect on the surface of the object to be inspected without erroneously detecting the harmless pattern even if the occurrence of the harmless pattern is changed.

DESCRIPTION OF SYMBOLS 1 Surface defect detection apparatus 2 Inspected object 2a Harmless pattern generation condition confirmation area 2b Welding part 3 Illumination part 4 Imaging part 5 Polarizer 6 Analyzer 7 Light source 8 Image input part 9 Image processing part 10 Image monitor 11 Polarization angle correction input part DESCRIPTION OF SYMBOLS 12 Polarization angle adjustment part 13 Feature-value calculation part 14 Defect determination part 15 External input / output part 16 Host computer 17 Light quantity control part 18 Coil 19 Winding coil 20 Image evaluation part

Claims (5)

An imaging means for receiving a reflected light of the illumination light applied to the object to be inspected, and imaging a surface image of the object to be inspected;
Changing means for changing an optical condition under which the imaging means performs imaging;
Evaluation means for evaluating the surface image of the object to be inspected, which is imaged by the imaging means;
Adjusting means for controlling the changing means according to the evaluation of the evaluating means;
A surface defect detection apparatus comprising: The imaging means includes a polarizer disposed in the optical path of the irradiation light and an analyzer disposed in the optical path of the reflected light,
The changing means changes the optical condition by changing a relative angle between the polarizer and the analyzer.
The surface defect detection apparatus according to claim 1. The evaluation means is an image monitor that displays a surface image of the object to be inspected,
The adjusting means is an input device for an operator to input the angle of the analyzer while checking the image monitor.
The surface defect detection apparatus according to claim 2. The evaluation means is an arithmetic unit that calculates a noise level of a surface image of the object to be inspected, and searches for an angle of the analyzer that minimizes the calculated noise level as a control angle,
The adjusting means controls the changing means so that an angle of the analyzer becomes the control angle;
The surface defect detection apparatus according to claim 2. The object to be inspected is a steel plate conveyed through a pickling line,
5. The surface defect detection apparatus according to claim 1, wherein the adjustment unit controls the changing unit at a timing at which a manufacturing condition of the steel sheet is switched.

Images