JP5347418B2 - Surface defect inspection system, method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enhance a detection accuracy of surface defects, even when an object to be measured is a long material, having a curve whose both ends are low or high in cross-sectional shape. <P>SOLUTION: A linear light extending in a wide direction of the object to be measured is irradiated on the object to be measured. The light reflected from the object to be measured is imaged by a TDI camera 30. Furthermore, an optical cutting image is acquired. In the surface defect detection inspection of the object to be measured, shape imaging which expresses the surface shape of the object to be measured is performed from the striped images which are acquired, by arranging in order the optical cutting images. After this procedure, a preliminary processing part 101 extracts a concavo-convex part on a shape image. Furthermore, a shape correction processing part 508 carries out fitting procedure so as to obtain a fitting curve for the shape image which concavo-convex part extracted by the preliminary processing part 101 from the shape image is removed, and performs a differential operation between the fitting curve and the shape image. A defect detection processing part 509 detects the surface defect of the object to be measured, based on the data obtained from the differential operation at the shape correction processing part 508. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention relates to a surface defect inspection system, method, and program suitable for use in detecting a surface defect using a long material having a curved line in which both end portions are lowered or raised in a cross-sectional shape.

  In recent years, a light-cutting optical shape measuring method has been used in various fields. Even in a steel plate production line, the surface of a steel plate surface, such as dents and wrinkles, is emitted by irradiating the surface of the steel plate with linear laser light, imaging the surface of the steel plate with a delay integration camera, and outputting a light cut image. It has been proposed to detect defects (see Patent Document 1).

JP 2004-3930 A Japanese Patent Laid-Open No. 2005-207858 JP 2008-2848 A

  In the above-mentioned Patent Document 1, it is considered that a steel plate having a flat surface is assumed as a measurement object. For example, when a measurement object is a rail, a shaped steel, a steel pipe, the surface is not a flat surface, Both ends are curved surfaces that become lower (or higher), and the size of surface defects can be larger or deeper than the width of the inspection surface. May become low.

  For example, when the measurement object is transported in the longitudinal direction, the measurement object may meander or vibrate in the width direction, but if the surface of the measurement object is a curved surface, the meandering Or vibration in the width direction may cause a slight shift at the end of the measurement object, which may be perceived as a large shape change (unevenness) in image processing (such as shading correction processing). . On the other hand, by incorporating a process (centering process) in which the left and right edges of the shape data are detected in advance and shifted to the left and right so that the center of the measurement object becomes the center of the image, the shape change that occurs in the process is reduced. . However, in particular, a defect of about 20 to 30 mm may occur in the inspection surface having a width of 100 mm or less particularly in the head portion or the head side portion of the rail. In this case, when the shading correction process is performed, the apparent shape is dragged to the defective part, and in the flattening process for detecting the defect, the irregularity and size of the defect are misidentified. In some cases, defects may be missed.

  Further, as shown in FIG. 25, it is conceivable to obtain the fitting curve 71 for each line of the shape image 72 after the centering process and flatten the curved surface shape, but the fitting is originally performed as shown in FIG. Where it should be, it is dragged to a defective part and fitting is performed as shown in FIG. In this case as well, the same bounce (unevenness generated in the process) 73 occurs in the width direction of the defect, and it is detected to be shallower and smaller than the original defect shape, which is overlooked. This oversight occurs even when the size of the defect is as small as 20 mm or less, or when the unevenness (depth or height) of the defect is large.

  Further, Patent Document 2 discloses a method for inspecting a steel surface defect in which a slit-like light beam is inclined and applied to a corner portion of a steel material, and an optical track formed thereby is imaged. In Patent Document 2, an image is binarized, a corner shape is approximated two-dimensionally using a least square method, a normal corner shape is estimated, and the estimated corner shape is compared with an actual corner shape. Whether or not there is a corner crack is determined based on the difference.

  However, even in Patent Document 2, when the ratio of the defect portion occupying the corner portion is large, an approximation result dragged to the defect portion is obtained as described above, and as a result, the unevenness and shape of the defect cannot be accurately recognized. .

  Patent Document 3 discloses a defect inspection method for a rod-like rotary tool for inspecting a rod-like rotary tool used for cutting or the like using image processing. In Patent Document 3, since the edge line of the cutting edge is photographed as an extremely simple shape such as a straight line or an arcuate curve on the image, it is assumed that the approximate line passing through the edge line is a normal shape, and the approximate line and the actual line A large deviation of the outline is detected as a defect. In order to fit the approximate line, a robust regression method in which a weight is added to the least square regression method is used.

  However, in Patent Document 3, an approximate line is obtained for a very simple shape such as a straight line or an arcuate curve. When the object to be measured is a rail, shape steel, steel pipe or the like whose surface is a curved surface, for example, if a fitting process using a higher order function of the fourth order or higher is performed, the surface defect of the object to be measured is large or deep. If this happens, the fitting curve may be dragged by this surface defect. In Patent Document 3, no countermeasure is taken into consideration when performing fitting processing using a higher-order function.

  The present invention has been made in view of the above points, and improves the detection accuracy of surface defects when a long material having a curved line whose both end portions are low or high in the cross-sectional shape is used as a measurement object. With the goal.

In the surface defect inspection system of the present invention, a long material having a curved line in which both end portions are lowered or raised in a cross-sectional shape is used as a measurement object, and a periodically modulated linear laser beam is applied to the surface of the measurement object. Using the irradiation means for irradiating in the width direction and the delay integration type imaging means for imaging the reflected light from the measurement object, the irradiation position of the linear laser light on the measurement object is continuously shifted. A surface defect inspection system for detecting a surface defect of the measurement object by imaging the reflected light from the measurement object by the imaging means and outputting a light cut image;
A phase calculation unit that calculates a phase shift at each position of the fringe image obtained by sequentially arranging the light-cut images obtained by the imaging unit, and a phase based on the phase shift obtained by the phase calculation unit Phase continuation that detects the position where the deviation of the measurement is discontinuous, and creates a shape image representing the surface shape of the measurement object by continuating the phase deviation by connecting the phase deviation at the detected position Processing means, pre-processing means for extracting irregularities on the shape image created by the phase continuation processing means, and a shape processing image extracted by the pre-processing means from the shape image created by the phase continuation processing means A fitting process for obtaining a fitting curve is performed on the shape image from which the uneven portion has been removed, and the difference between the fitting curve and the shape image created by the phase continuation processing means Calculates have rows, the data obtained in the difference calculation, the average in the longitudinal direction by a predetermined length of the measurement object, and a value obtained by performing a low-pass filter processing on the average value, by the difference calculation and again row cormorant shape correcting means the difference calculation between data obtained on the basis of the data obtained by the difference calculation of the shape correction process unit, and the defect detection processing means for detecting a surface defect of the object to be measured It is provided with.
Another feature of the surface defect inspection system according to the present invention is that the preprocessing means performs a fitting process for obtaining a fitting curve on the shape image created by the phase continuation processing means, A primary shape correction processing means for performing a difference calculation between the fitting curve and the shape image created by the phase continuation processing means, and a binary value for binarizing data obtained by the difference calculation of the primary shape correction processing means A concavo-convex portion extracting means for extracting the concavo-convex portion from the binarized image binarized by the binarizing means, and the concavo-convex portion extracting means from the shape image created by the phase continuation processing means. And a mask processing means for generating mask information for removing the extracted uneven portion.
Another feature of the surface defect inspection system according to the present invention is that the pre-processing means averages the shape image created by the phase continuation processing means in the longitudinal direction of the measurement object. A primary shape correction processing unit that creates a reference shape image and performs a difference calculation between the reference shape image and the shape image generated by the phase continuation processing unit, and a difference calculation between the primary shape correction processing unit. Created by the binarization means for binarizing the obtained data, the unevenness extracting means for extracting the unevenness from the binarized image binarized by the binarization means, and the phase continuation processing means And a mask processing means for generating mask information for removing the uneven portions extracted by the uneven portion extracting means from the shape image.
Further, another feature of the surface defect inspection system of the present invention is that expansion is performed for expanding the uneven portions extracted by the uneven portion extracting means at least in a predetermined ratio in the width direction of the measurement object. And a mask processing unit for generating mask information for removing the concavo-convex portion expanded by the expansion processing unit from the shape image generated by the phase continuation processing unit.
According to the surface defect inspection method of the present invention, a long material having a curved line whose both end portions are low or high in a cross-sectional shape is used as a measurement object, and periodically modulated linear laser light is applied to the surface of the measurement object. Using the irradiation means for irradiating in the width direction and the delay integration type imaging means for imaging the reflected light from the measurement object, the irradiation position of the linear laser light on the measurement object is continuously shifted. A surface defect inspection method for detecting a surface defect of the measurement object by imaging reflected light from the measurement object by the imaging means and outputting a light-cut image, and the light obtained by the imaging means A phase calculation procedure for calculating a phase shift at each position of the fringe image obtained by sequentially arranging the cut images, and a phase shift becomes discontinuous based on the phase shift obtained by the phase calculation procedure. Have A phase continuation processing procedure for creating a shape image representing the surface shape of the measurement object by detecting a position and continuating the phase lag by connecting the phase lag at the detected position, and the phase continuation processing A pre-processing procedure for extracting irregularities on the shape image created in the procedure, and a shape image obtained by removing the irregularities extracted in the pre-processing procedure from the shape image created in the phase continuation processing procedure Te subjected to fitting processing for obtaining the fitting curve, and its fitting curve, have rows difference operation between shape images created by the phase continuous process procedure, the data obtained in the difference calculation, the measurement object An average of a predetermined length in the longitudinal direction, and a difference calculation between the value obtained by performing the low-pass filter process on the average value and the data obtained by the difference calculation is performed again. Cormorants and shape correction procedure, the shape correction procedure based on data obtained by the difference calculation of characterized by having a defect detection procedure for detecting a surface defect of the object to be measured.
The program of the present invention uses a long material having a curved line whose both end portions are low or high in the cross-sectional shape as a measurement object, and periodically modulates a linear laser beam in the width direction of the surface of the measurement object. Using the irradiating means for irradiating and the delay integration type imaging means for imaging the reflected light from the measurement object, the imaging position is continuously shifted while the irradiation position of the linear laser light on the measurement object is shifted. Means for capturing reflected light from the measurement object and outputting a light-cut image, and detecting a surface defect of the measurement object, the light-cut images obtained by the imaging means in order A phase calculation process for calculating a phase shift at each position of the fringe image obtained by arranging, and a position where the phase shift is discontinuous based on the phase shift obtained by the phase calculation process. A phase continuation process for creating a shape image representing the surface shape of the measurement object by continuating the phase shift by connecting the phase shifts at the detected position, and the phase continuation process Pre-processing for extracting irregularities on a shape image, and fitting processing for obtaining a fitting curve for a shape image obtained by removing the irregularities extracted by the pre-processing from the shape image created by the phase continuation processing subjected, its fitting curve, have rows difference operation between shape image generated in the phase-continuous process, the average data obtained by the difference calculation, a predetermined length in the longitudinal direction of the measurement target to, the value obtained by performing a low-pass filter processing to a value obtained by averaging, and again a row cormorant shape correction process the difference calculation between the obtained data in the difference calculation, said shape complement Based on the data obtained by the difference calculation processing, to perform a defect detection process for detecting surface defects of the object to be measured to the computer.

  ADVANTAGE OF THE INVENTION According to this invention, the detection accuracy of a surface defect can be improved when the length object which has a curve which both ends become low or becomes high in a cross-sectional shape is used as a measuring object. That is, in the present invention, the fitting curve is obtained in the form of removing the uneven portion from the shape image representing the surface shape of the measurement object, and the difference calculation between the fitting curve and the shape image is performed. It is possible to prevent the fitting curve from being dragged. As a result, the surface defects are small and can be prevented from being detected shallowly, and the occurrence of rebounding on both sides thereof can be prevented.

  Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. First, with reference to FIGS. 1 to 14, a technique that is a premise of a surface defect inspection method according to an embodiment to which the present invention is applied will be described as a comparative example.

(Comparison)
FIG. 1 is a schematic configuration diagram of a surface defect inspection system, in which a laser device 10, a rod lens 20, a delay integration camera (TDI camera) 30, a timing signal generator 40, an image processing device 50, and a display device are illustrated. 60.

  Such a surface defect inspection system optically measures the surface shape of the measuring object 2. Here, as the measuring object 2, as shown in FIG. 2, a long material (long plate or bar) having a curve in which both ends are lowered (or increased) in the cross-sectional shape, that is, the surface is both ends. Are assumed to be long materials having curved surfaces that become lower (or higher), such as rails, shaped steels, steel pipes, and the like. For example, assume that the surface has a height difference of about 6 mm. The measurement object 2 is conveyed at a constant speed (for example, 60 mpm) in the longitudinal direction (arrow in the figure), and the surface defect inspection system of this example measures the surface shape of the measurement object 2 during conveyance. Then, surface defects such as dents and wrinkles on the surface of the measuring object 2 are detected.

  The laser device 10 generates a continuous wave laser beam. The rod lens 20 spreads the laser beam emitted from the laser device 10 in a fan shape along the width direction of the measurement object 2 (direction perpendicular to the paper surface of FIG. 1). Thereby, the laser beam emitted from the laser device 10 is irradiated onto the measurement object 2 as a linear laser beam. For example, a laser device with an output of 500 mW and a wavelength of 670 nm is used to irradiate the surface of the measuring object 2 as 400 mm × 0.2 mm linear laser light. At this time, the linear laser light is incident obliquely (for example, 45 degrees) with respect to the surface of the measurement object 2. Thus, on the surface of the measuring object 2 irradiated with the linear laser beam, a linear bright part is formed along the width direction of the measuring object 2. Further, since the measurement object 2 moves in the longitudinal direction, when viewed from the measurement object 2, the linear bright part also moves along the longitudinal direction of the measurement object 2. The reflected light (linear reflected image) from such a linear bright part is picked up by the TDI camera 30.

  The timing signal generator 40 generates a sine waveform signal having a predetermined frequency ω (for example, 5 kHz) and sends the sine waveform signal to the laser device 10. The laser device 10 can continuously change its oscillation intensity by an external signal, and generates a laser beam whose output changes in a sine waveform when receiving a sine waveform signal sent from the timing signal generator 40. To do. That is, the laser beam emitted from the laser device 10 is periodically modulated. The timing signal generator 40 generates a camera shift pulse signal having a frequency M times (for example, 4 times) the frequency ω, and sends the camera shift pulse signal to the delay integration type camera 30.

  The TDI camera 30 captures a linear reflection image of the moving measurement object 2. For example, a TDI camera having a resolution of 0.2 mm × 0.2 mm is used, and the TDI camera is arranged in a direction substantially perpendicular to the surface of the measurement object 2. FIG. 3 is a diagram for explaining the structure and operation of the TDI camera. In the TDI camera 30, as shown in FIG. 3A, a large number of photoelectric conversion elements 35 are arranged in a matrix. Here, for example, it is assumed that n photoelectric conversion elements 35 are arranged in the row direction and N in the column direction. Each row is numbered sequentially from the top with the top row as the first row, and each column is numbered sequentially from the left with the leftmost column as the first column. Here, the row direction of the photoelectric conversion elements 35 corresponds to the width direction of the measurement object.

  The photoelectric conversion element 35 accumulates charges corresponding to the intensity of the received light. It is assumed that the linear reflection image of the measurement object 2 is incident on the photoelectric conversion element 35 with a width corresponding to one row through the lens 31 of the TDI camera 30. In the TDI camera 30, each photoelectric conversion element 35 transfers the accumulated charge to a photoelectric conversion element located in the same row as the photoelectric conversion element 35 and located in the next column. The timing of this transfer is the same for all the photoelectric conversion elements 35, and is controlled by a camera shift pulse signal sent from the timing signal generator 40. That is, each time a camera shift pulse signal is input, each photoelectric conversion element 35 transfers charges. The frequency of the camera shift pulse signal (camera shift frequency) is Mω. Then, when the camera shift pulse signal is input, the photoelectric conversion element 35 located in the Nth column (final column) sends the accumulated charge to the read register. Thereby, the light section image corresponding to the linear reflection image is output.

  In general, in the TDI camera 30, as shown in FIG. 3B, when light is incident on each photoelectric conversion element 35 while the charge is being transferred, a charge corresponding to the intensity of the incident light is added. Is done. However, as described above, a linear reflection image having a width corresponding to one column is incident on the photoelectric conversion element 35. For this reason, the charge is hardly added in each photoelectric conversion element 35 during the transfer of the charge. In particular, even when background light becomes a problem, this can be suppressed by installing an optical filter that transmits only the laser wavelength in front of the lens.

  Since the measuring object 2 moves along the longitudinal direction, the laser beam is irradiated from the laser device 10 onto the measuring object 2, and a linear reflection image of the measuring object 2 is used for a predetermined time using the TDI camera 30. When the image is taken, it is possible to sequentially obtain light-cut images at each position in the longitudinal direction of the measuring object 2. Therefore, an image representing the entire measurement object 2 can be obtained by sequentially arranging the light section images thus obtained.

  Further, the linear laser light is periodically modulated, and the intensity of the linear laser light changes with time, so that the amount of charge (received light intensity) accumulated in each photoelectric conversion element in the column direction in each row. The distribution also changes periodically. Therefore, an image obtained by sequentially arranging the light section images output from the TDI camera 30 becomes a fringe image in which the density (intensity) of each light section image periodically changes along the array direction. . FIG. 4 shows an example of a stripe image. Here, the light section image corresponding to one period of density change is referred to as “stripe”. In such a striped image, the direction parallel to the stripe corresponds to the width direction of the measurement object 2, and the direction orthogonal to the stripe corresponds to the longitudinal direction of the measurement object 2. When the ratio of the camera shift frequency of the TDI camera 30 to the modulation frequency of the laser light is M: 1, M light-cut images, that is, M pixels in the arrangement direction form one stripe.

  By the way, since the laser beam is incident on the surface of the measurement object 2 at an angle, for example, if there is a recessed portion in the measurement object 2, the reflection point of the laser beam is shifted to the right in FIG. Therefore, the position of the light section image on the photoelectric conversion element 35 is also shifted in the column direction in the right side, that is, in FIG. For this reason, in the fringe image, the light cut image corresponding to the laser light reflected by the concave portion is output earlier in time than the light cut image corresponding to the laser light reflected by the non-dented portion. It will be. Therefore, in the image obtained by arranging the images output from the TDI camera 30 in order, the recessed portion can be clearly recognized as a stripe shift.

  This fringe shift will be described in more detail. FIG. 5A is a schematic enlarged view of a certain fringe image. In FIG. 5A, the position where the maximum density is given for each stripe is shown by a solid line. For example, in this fringe image, when the maximum density position is examined along the arrangement direction at the position A in the width direction, the maximum density positions are located at equal intervals, and no stripe deviation occurs. That is, the measurement object 2 has a flat shape at the position A in the width direction along the arrangement direction. In this case, the density distribution of the fringe image (slice fringe image data) along the arrangement direction at the position A in the width direction has a clean sine wave shape as shown in FIG.

  On the other hand, when the maximum density position is examined along the arrangement direction at the position B in the width direction shown in FIG. 5A, the interval between the maximum density positions gradually widens from the left to the right, and a stripe shift occurs. ing. That is, the measurement object 2 has a dent along the arrangement direction at the position B in the width direction. In this case, the density distribution (slice fringe image data) of the fringe image along the arrangement direction at the position B in the width direction is compared with the sine wave shown in FIG. 5B as shown in FIG. Out of phase. As described above, the fringe shift due to the dent of the measurement object 2 appears as a phase shift in the slice fringe image data. Actually, the phase shift and the dent (depth) of the measuring object 2 are in a proportional relationship. The greater the depth, the greater the phase shift in the slice fringe image data. In the surface defect inspection system, information related to the phase shift is calculated based on the fringe image, and the surface shape of the measurement object 2 is measured based on the information related to the phase shift. If this method of calculating the phase shift is used, for example, when the ratio of the camera shift frequency to the modulation frequency of the laser beam is set to 4: 1, the fringes can be detected with unevenness even if the fringes are not clearly shifted by one pixel or more. There is sensitivity.

  Here, the fringe image shown in FIG. 4 is that of a measurement object having a flat surface. On the other hand, the measurement object 2 assumed in this example, that is, a long material having a curve in which the both end portions are lowered (or increased) in the cross-sectional shape such as a rail, a shape steel, a steel pipe, etc. is shown in FIG. Thus, it becomes a fringe image in which both ends of each fringe are curved.

  Next, the relationship between the phase shift in the slice fringe image data and the depth of the measurement object 2 will be described. FIG. 7 is a diagram for explaining the relationship between the phase shift in the slice fringe image data and the depth of the measurement object 2.

  Now, as shown in FIG. 7, the incident angle at which the linear laser light is incident on the surface of the measuring object 2 is defined as θ. Further, it is assumed that the measurement object 2 has a recess, and the linear laser light is reflected at a depth d from the surface of the measurement object 2 and enters the TDI camera 30 when entering the recess. At this time, the linear laser beam reflected at the depth d is reflected by the distance h in the longitudinal direction (right direction) of the measuring object 2 as compared with the linear laser light reflected by the flat surface of the measuring object 2. The point shifts. Here, h = d · tan θ. As a result of the deviation of the reflection point of the linear laser beam by the distance h in the longitudinal direction, a phase shift occurs in the slice fringe image data. This phase shift is denoted by φ.

Assuming that the imaging resolution in the column direction of the photoelectric conversion element 35 in the TDI camera 30 is s (mm / pixel), the distance h that the reflection point of the linear laser beam is displaced in the longitudinal direction corresponds to the h / s pixel in the striped image. To do. When the ratio between the camera shift frequency of the TDI camera 30 and the modulation frequency of the laser beam is M: 1, M pixels in the arrangement direction form one stripe in the stripe image. That is, when the fringes are shifted by M pixels, the phase shift is 2π. Therefore, the phase shift φ in the slice fringe image data when the reflection point of the linear laser beam is shifted by the distance h in the longitudinal direction is
M / 2π = (h / s) / φ
Than,
d = {M · s / (2π · tan θ)} φ
It becomes. From this, it can be seen that the phase shift φ in the slice fringe image data and the depth d of the measurement object 2 are in a proportional relationship.

  Strictly speaking, when a normal lens is used, the imaging resolution s changes according to the depth d and needs to be corrected. However, as in the case of measuring a dent in a steel plate, the lens working distance is not affected. When the depth change is small, such a change in the imaging resolution s can be ignored in practice. If a telecentric lens is used, the imaging resolution s can be made constant regardless of the depth d.

  The image processing device 50 performs processing for creating an image representing the surface shape of the measurement object 2 based on each light cut image output from the TDI camera 30, and processing for detecting a defect based on the image. It is. FIG. 8 shows a schematic block diagram of the image processing apparatus 50. The image processing apparatus 50 includes an A / D conversion unit 501, a pre-filter unit 502, an orthogonal sine wave generation unit 503, low-pass filter units 504a and 504b, a phase calculation unit 505, a phase continuation processing unit 506, A centering processing unit 507, a shape correction processing unit 508, a defect detection processing unit 509, an amplitude calculation unit 510, and an edge position detection unit 511 are included. The results processed by each part of the image processing device 50 are displayed on the screen of the display device 60.

  FIG. 14 is a flowchart for explaining the processing operation of the image processing apparatus 50. Hereinafter, the processing operation of each unit of the image processing apparatus 50 will be described with reference to FIG.

  First, the TDI camera 30 images the measurement object 2 and outputs a light cut image (step S101). The A / D conversion unit 501 performs A / D conversion on each light-cut image output from the TDI camera 30 and outputs it as digital multilevel image data. Such digital multilevel image data is stored in an image memory (not shown). By arranging these digital multivalued image data in order, a fringe image is formed.

  From such a fringe image (or digital multi-valued image data), data representing the density distribution of the fringe image along the arrangement direction at each position in the width direction is generated. Data representing the density distribution of the stripe image along the arrangement direction is “slice stripe image data”. Slice fringe image data at each position in the width direction is sequentially output from the image memory.

  The pre-filter unit 502 removes noise and sharpens the stripe state by applying a predetermined filter process to each slice stripe image data. Note that the filtering process by the prefilter unit 502 is not necessarily performed. For example, it may be performed only when many fine noises are generated in the striped image.

Two pieces of slice fringe image data I _j (k) at each position j (j = 0, 1, 2,...) In the width direction are output from the prefilter unit 502. k (k = 0, 1, 2,...) is a position in the arrangement direction. At this time, it is assumed that the slice fringe image data I _j (k) at the position j in the width direction changes sinusoidally. That is,
I _j (k) = A (j, k) {cos ((2πk / M) + φ (j, k)) + 1}
It is. Here, A (j, k) is the amplitude of the slice fringe image data at the pixel position (j, k), and φ (j, k) is the phase shift of the slice fringe image data at the pixel position (j, k). . The influence of the fringe shift generated in the fringe image due to the depression of the measurement object 2 appears as a phase shift φ. Since the amplitude of the linear laser beam is constant, the amplitude A is usually constant. However, when the surface of the measuring object 2 is dirty, the amplitude A may decrease rapidly at the pixel position corresponding to the dirty position. Therefore, in the above equation, the amplitude A is written in a form depending on the pixel position (j, k).

The reason why “1” is added after the term of cos is to guarantee this because slice stripe image data (density value) I _j (k) does not become negative. Therefore, the slice fringe image data I _j (k) varies between 0 and 2A.

The orthogonal sine wave generation unit 503 generates two orthogonal reference sine wave data sin (2πk / M) and cos (2πk / M), which are created in advance on a memory such as a ROM. In particular, the former is also referred to as reference sin data, and the latter is also referred to as reference cos data. These two reference sine wave data are respectively multiplied by the slice fringe image data I _j (k) output from the prefilter unit 502. By this multiplication processing, the following two outputs Ia _j (k) and Ib _j (k) are obtained.

Each of the low-pass filter units 504a and 504b performs predetermined filter processing on the outputs Ia _j (k) and Ib _j (k) obtained by the above multiplication processing, thereby removing the fringe frequency component and its harmonic components. That is, a component including only the phase shift φ is extracted. If the output from the low-pass filter unit 504a is LPF (Ia _j (k)) and the output from the low-pass filter unit 504b is LPF (Ib _j (k)),
LPF (Ia _j (k)) = (A cos φ) / 2
LPF (Ib _j (k)) = − (A sin φ) / 2
It is.

  The phase calculation unit 505 calculates a phase shift φ (j, k) at each pixel position (j, k) based on the results output from the two low-pass filter units 504a and 504b (step S102). The phase shift φ (j, k) can be obtained from the following equation.

In the above equation, the arctan value range is set to −π / 2 to + π / 2, and information about the codes of LPF (Ia _j (k)) and LPF (Ib _j (k)) is used to shift the phase. φ is determined in the range of −π to + π. Here, the phase shift obtained in this range will be described again as φ ′. In this case, the phase shift φ ′ obtained by the above formula has a periodic relationship with the depth of the measurement object 2, and there are a plurality of depths that take a certain value of the phase shift φ ′. Therefore, accurate information on the surface shape of the measuring object 2 cannot be obtained by using such a phase shift φ ′. For this reason, it is necessary to obtain a phase shift φ that is proportional to the depth of the measurement object 2 from the phase shift φ ′. The process of obtaining a phase shift φ proportional to the depth is performed by the phase continuation processing unit 506. FIG. 9A shows an example of a phase image created by the phase calculation unit 505. FIG. 10A shows an example of a characteristic diagram representing the phase characteristic in the phase image.

  The phase continuation processing unit 506 detects discontinuous points of the phase shift φ ′ based on the phase image created by the phase calculation unit 505, and the phase shift φ ′ so that the phase shift φ ′ is smoothly connected. 'Is corrected (step S103). As described above, since the value range of the phase shift φ ′ calculated by the phase calculation unit 55 is −π to + π, the phase shift φ ′ is discontinuous at −π and + π. For example, in the phase image shown in FIG. 9A, a portion that changes from white (or black) to black (or white) corresponds to a discontinuous point of the phase shift φ ′. If such a phase image is used as it is, it is difficult to recognize the surface shape of the measuring object 2. Therefore, it is necessary to correct the phase shift φ ′ so that the phase shift φ ′ is smoothly connected at the discontinuous points of the phase shift φ ′. Such correction (phase jump correction) is a process for obtaining a unique phase shift φ proportional to the depth of the measuring object 2 from the phase shift φ ′ defined in the range of 2π.

  Specifically, the phase continuation processing unit 506 detects a discontinuous point of the phase shift φ ′ and corrects the phase shift φ ′ at the discontinuous point. Whether or not the phase shift φ ′ is discontinuous is not known by looking at only one pixel. It is necessary to judge by looking at adjacent pixels. First, the phase continuation processing unit 506 checks the phase image along the arrangement direction at each position in the width direction of the phase image, and compares the phase shift φ ′ between adjacent pixels. When the phase shift φ ′ is greatly different between the adjacent pixels, it is determined that the phase shift φ ′ is discontinuous between the pixels, and the phase shift φ ′ is corrected. Actually, the depth at the surface of the measuring object 2 such as a steel plate does not change abruptly. For this reason, it is considered that the phase shift φ ′ is largely different because the phase shift φ ′ is changed by ± 2π. Therefore, a pixel whose phase shift φ ′ is significantly different from the phase shift φ ′ of the adjacent pixels is examined, and the phase shift φ ′ may be smoothly connected (FIG. 10). (See arrow in (a)).

  For example, when a phase shift φ ′ is close to + π at a certain pixel position and the phase shift φ ′ is close to −π at a pixel position on the right side, the right adjacent pixel At the position, it is recognized that the phase shift φ ′ changes by + 2π. Then, the phase shift φ ′ is corrected by adding + 2π to the phase shift φ ′ at the right adjacent pixel position. Further, when the phase shift φ ′ is a value close to −π at a certain pixel position, and the phase shift φ ′ is a value close to + π at the pixel position on the right side, the pixel on the right side At the position, it is recognized that the phase shift φ ′ changes by −2π. The phase shift φ ′ is corrected by adding −2π to the phase shift φ ′ at the right pixel position.

  Thus, after examining the adjacent pixels along the arrangement direction at each position in the width direction and correcting the phase shift φ ′, the phase continuation processing unit 506 is now arranged in the width direction at each position in the arrangement direction. The adjacent pixels are examined, and similarly, the phase shift φ ′ is corrected. The phase shift at each pixel position after correction is a unique phase shift φ proportional to the depth of the measurement object 2.

  Next, the phase continuation processing unit 506 newly creates a phase image based on the corrected phase shift φ. This new phase image accurately represents the surface shape of the measuring object 2. This new phase image is referred to as a shape image. FIG. 9B shows an example of a shape image created by the phase continuation processing unit 506. FIG. 10B shows an example of a characteristic diagram representing the phase characteristic in the shape image.

On the other hand, the amplitude calculation unit 510 calculates the amplitude A (j, k), that is, the luminance at each pixel position (j, k) based on the results output from the two low-pass filter units 504a and 504b (step S104). ). The amplitude A (j, k) is
A (j, k) = 2 [{LPF (Ib _j (k))} ² + {LPF (Ia _j (k))} ² ] ^1/2
It can be obtained more. Then, an amplitude image (luminance image) is created based on the calculated amplitude A. The amplitude image is expressed as a grayscale image in which the image becomes black as the amplitude is small, for example. FIG. 11 shows an example of an amplitude image created by the amplitude calculation unit 510.

  The edge position detection unit 511 detects the left and right edge positions for each line based on the amplitude image (luminance image) created by the amplitude calculation unit 510 by using a threshold value (step S105).

  The centering processing unit 507 performs centering processing on the shape image created by the phase continuation processing unit 506 based on the edge position detected by the edge position detection unit 511 (step S106). When photographing the measurement object conveyed in the longitudinal direction, if the measurement object meanders or vibrates, the image is shifted in the width direction as shown in FIGS. 12 (a) and 12 (b). Taken. FIG. 12A is a schematic diagram showing a state when the measurement target is meandering, and FIG. 12B is a schematic diagram showing a state when the measurement target is vibrated, and the solid line is the left and right edges of the measurement target. The dotted line indicates the center position.

  For such an image, in the centering process, the left and right edge positions of the inspection surface are detected for each line, and the centers thereof are calculated as the center positions. This position is aligned with the image center in the vertical direction. FIG. 12C is a schematic diagram showing an image after the centering process. Due to the centering process, when the measurement object 2 is a long material having a curved line whose both end portions are low or high in the cross-sectional shape, the measurement object 2 meanders or vibrates in the width direction. Sometimes, the detection accuracy of surface defects can be increased. In this case, the image used for edge position detection may be a shape image or a luminance image.

  The shape correction processing unit 508 performs shape correction processing on the shape image after the centering processing. In the shape images shown in FIGS. 9 (b) and 10 (b), the surface shape of the measurement object 2 itself is not flat. Therefore, even if the binarization process is performed to determine the surface defect, the surface of the steel plate is recessed. It is impossible to extract surface defects such as rust and wrinkles. Therefore, a shape correction process is performed in which the curved surface shape of the measuring object 2 is baked into a flat surface.

Specifically, the shape image after the centering process is subjected to a fitting process using a high-order function (for example, an even function of 4th order or higher) for each line (step S107). The fitting process is a process for obtaining a curve (fitting curve) having the smallest residual square sum for a given data string, and a general least square method, a Chebyshev method, or the like can be used. The least square method is a process of calculating a coefficient by setting a partial differential coefficient based on a residual sum of squares coefficient to zero using a power of a variable as a basis function. At this time, if a Chebyshev polynomial, which is an orthogonal polynomial, is used as the basis function, a higher-order function can be fitted more accurately and stably. A fitting function based on a Chebyshev polynomial is expressed by the following equation, and coefficients a _i (i = 0, 1,..., N + 1) are obtained by fitting processing.

  Subsequently, a difference between the fitting curve generated in the fitting process (step S107) and the original data (shape image after the centering process) is calculated for each line (step S108). In FIG. 10B, the dotted line is the fitting curve, and the solid line is the original data. By this fitting process, even when both ends of the measurement object 2 have a steep shape, a sharp signal change at both ends is alleviated.

  Further, the data obtained by the difference calculation (step S108) is averaged (integrated) for a predetermined length in the traveling direction (longitudinal direction) of the measuring object 2 (step S109), and low-pass filter processing is performed (step S110). ), Difference operation (or division operation) with the original data (data obtained by difference operation (step S108)) is performed (step S111). By this shading correction, flattening is performed while retaining unevenness information that is a surface defect candidate.

  As described above, as the shape correction process, a two-stage process such as a fitting process using a high-order function (for example, a quadratic or higher-order even function) and a shading correction is performed. The image after the shape correction process is referred to as a depth image. FIG. 9C shows an example of a depth image created by the shape correction processing unit 508. FIG. 10C shows an example of a characteristic diagram representing the phase characteristic in the depth image.

  The defect detection processing unit 509 binarizes the depth image created by the shape correction processing unit 508 (step S112). In the binarization processing, a threshold value is set, and “0” is given to the portion within the threshold value, and “1” is given to the portion exceeding the threshold value, and imaged.

  Then, surface defect determination is performed using the binarized image (step S113), and the surface defect is detected (step S114). FIG. 13A shows a photograph of a binarized image. As shown in FIG. 13A, the number of pixels of the “1” portion of the binarized image is calculated and multiplied by the imaging resolution to calculate the defect area, length, width, aspect ratio, and the like. What is necessary is just to perform surface defect discrimination. In addition to a binarized image, a depth image or an amplitude image (luminance image) may be used. FIG. 13B is a diagram illustrating a photograph of a depth image, and FIG. 13C is a diagram illustrating a photograph of a luminance image. As shown in FIG. 13B, the surface defect is determined based on the maximum and minimum depths of the depth image, or the amplitude generated by the amplitude calculation unit 510 as shown in FIG. Surface defect discrimination may be performed based on the maximum luminance, minimum luminance, average luminance, and the like of an image (luminance image).

(First embodiment)
In the above-described comparison, the fitting process (step S107) in the shape correction processing unit 508 is performed on the shape image after the centering process. In this case, as shown in FIG. If the surface defect is large or deep, the fitting curve 71 may be dragged to the surface defect. For this reason, when the difference calculation with the original data (shape image after the centering process) 72 is performed, the surface defect is detected to be smaller and shallower than the actual one, and a bounce 73 occurs on both sides thereof. In FIG. 17A, two straight lines 74 indicated by dotted lines represent threshold values for binarization. When both convex and concave defects are to be detected, two threshold values indicating the upper and lower limits are set as described above, and the portion exceeding the threshold value in any of the concave and convex directions with respect to the shape level of the healthy part Is “1”.

  Therefore, in the embodiment to which the present invention is applied, as shown in FIG. 15, a pre-processing unit 101 is provided before the shape correction processing unit 508, and prior to the fitting processing in the shape correction processing unit 508, the post-centering processing is performed. Pre-processing for extracting irregularities on the shape image is performed. In addition, the same code | symbol is attached | subjected to the component already demonstrated in FIG. 8, and the detailed description is abbreviate | omitted.

  FIG. 16 is a flowchart for explaining the processing operation of the image processing apparatus 50 according to the present embodiment. The processes already described in FIG. 14 are denoted by the same reference numerals and detailed description thereof is omitted.

  After the centering process (step S106), the preprocessing unit 101 performs a primary shape correction process (step S201). Specifically, as shown in FIG. 18, first, a fitting process using a higher-order function (for example, an even or higher-order function) is performed for each line on the shape image after the centering process. As described above, the fitting process is a process for obtaining a curve (fitting curve) having the smallest residual sum of squares for a given data string. Then, for each line, a difference calculation between the fitting curve generated by the fitting process and the original data (the shape image after the centering process) is performed.

  Next, the data obtained by the difference calculation in the primary shape correction process in step S201 is binarized (step S202). In the binarization processing, a threshold value is set, and “0” is given to the portion within the threshold value, and “1” is given to the portion exceeding the threshold value, and imaged. FIG. 19A shows an example of a binarized image obtained by binarization processing.

  Next, an uneven part is extracted using the binarized image obtained by the binarization process of step S202 (step S203). An uneven part is extracted based on a part (pixel group to which “1” is given) exceeding the threshold value on the binarized image.

  Next, an expansion process is performed to enlarge the uneven part extracted by the uneven part extracting process in step S203 at a predetermined rate in the width direction of the measurement object 2 (step S204). In the primary shape correction process (fitting process) in step S201, as described in FIG. 17A, when the unevenness of the measurement object 2 is particularly large or deep, the fitting curve 71 is dragged to the uneven part. Therefore, when the difference calculation with the original data (shape image after the centering process) 72 is performed, the uneven portion is detected to be smaller and shallower than the actual size, and a bounce 73 may occur on both sides thereof. In view of this, the extracted uneven portion is enlarged and handled at a predetermined rate in the width direction of the measurement object 2.

  In the expansion processing, as shown in FIG. 20A, when each line is scanned and there is a pixel group to which “1” is given (a portion extracted as an uneven portion), both sides in the width direction thereof The pixel of “0” is changed to “1”. The number of pixels on both sides to be changed from “0” to “1” is determined in advance according to the number of pixels to which “1” is given. FIG. 19B shows an example of a binarized image obtained by the expansion process.

  In this embodiment, the example in which the measurement object 2 is expanded only in the width direction has been described. However, as shown in FIG. 20B, not only in the width direction but also in the two-dimensional direction (width direction and longitudinal direction). It does not matter if it is inflated. For example, as shown in FIG. 20B, surrounding “0” pixels may be changed to “1”.

  Then, the shape image after the centering process is matched with the concavo-convex portion information (hereinafter referred to as mask information) extracted and expanded in steps S203 and S204, and “0” in the mask information in the shape image after the centering process. A mask process is performed in which only the part corresponding to the pixel to which the value is given is calculated by the shape correction processing unit 508 (step S205). As a result, a shape image is created by removing the concavo-convex portion from the shape image after the centering process based on the mask information.

  In the shape correction processing unit 508 subsequent to the preprocessing unit 101, the shape image (the shape image obtained by removing the uneven portion based on the mask information from the shape image after the centering process) subjected to the mask processing in the preprocessing unit 101 is already stored. The fitting process described above is performed (step S107), and a difference calculation between the fitting curve generated by the fitting process and the original data (the shape image after the centering process) is performed (step S108).

  As described above, the fitting process (step S107) in the shape correction processing unit 508 is performed on the shape image (the uneven part based on the mask information from the shape image after the centering process) on which the mask processing is performed in the pre-processing unit 101. This is performed on the removed shape image). In this case, as shown in FIG. 17B, the surface defect of the measuring object 2 can be masked, and the fitting curve 71 can be prevented from being dragged by the surface defect. As a result, when the difference calculation with the original data (the shape image after the centering process) 72 is performed, the surface defect is small and avoids being detected shallowly, and bounce that does not actually exist on both sides of the surface defect occurs. Can be prevented.

  FIG. 21 shows a comparative detection result, and FIG. 22 shows a detection result in the present embodiment. FIG. 21A and FIG. 22A schematically show actually obtained images. 21B and 22B, A is the shape data after the centering process, B is the fitting curve, and C is the difference between them. In FIG. 21, it can be seen that the surface defects are detected smaller and shallower than the actual ones, and bounce occurs on both sides thereof. On the other hand, in FIG. 22, it can be seen that the surface defect is accurately detected reflecting the actual shape, and that no rebounding occurs on both sides thereof.

  FIG. 23 is a diagram showing the actual value (depth) of the examiner, the proportionality and the result (depth) of the surface defect inspection according to the present embodiment. ○ is a comparative result, and ● is a result in this embodiment. FIG. 23 also shows that the result of the surface defect inspection based on the proportionality is shallower than the actual surface defect, especially for the surface defect of a large area, whereas the result of the surface defect inspection according to the present embodiment is actually It can be seen that this is close to the surface defect.

  In the present embodiment, it has been described that the pre-processing unit 101 performs the process up to removing the concavo-convex part from the shape image after the centering process based on the mask information. However, the pre-processing unit 101 includes the process up to generating the mask information. Then, the mask information may be passed to the shape correction processing unit 508, and the shape correction processing unit 508 may perform a process of removing the concavo-convex portion from the shape image after the centering process based on the mask information.

(Second Embodiment)
The second embodiment is an example in which the content of the primary shape correction process in the preprocessing unit 101 is changed. Note that the configuration and processing operation of the image processing apparatus 50 are the same as those in the first embodiment, and the description thereof will be omitted below, and differences from the first embodiment will be mainly described.

  After the centering process (step S106), the preprocessing unit 101 performs a primary shape correction process (step S201). Specifically, as shown in FIG. 24, first, the reference shape image 75 is created by averaging the values of the respective lines of the shape image 72 after the centering process in the longitudinal direction of the measuring object 2. Then, the difference calculation between the reference shape image 75 and the original data (shape image after the centering process) 72 is performed for each line.

  Subsequent binarization processing (step S202), concavo-convex portion extraction processing (step S203), expansion processing (step S204), and mask processing (step S205) are as described in the first embodiment.

  When the reference shape image 75 is created by integrating and averaging in the longitudinal direction of the measurement object 2, as in the case of the fitting process, if the unevenness of the measurement object 2 is particularly large or deep, a reference is made to the uneven portion. The shape image 75 may be dragged (distortion occurs in the reference shape image 75). For this reason, when the difference calculation with the original data (shape image after the centering process) 72 is performed, the uneven portion is detected to be small and shallow. Therefore, in the expansion process (step S204), the extracted uneven part is enlarged and handled at a predetermined rate in the width direction of the measurement object 2.

  The “shape image created by the phase continuation processing means” in the present invention includes the shape image itself created by the phase continuation processing unit 506 and the shape image itself subjected to some processing. Shall. In the first and second embodiments, the centering process is performed on the shape image created by the phase continuation processing unit 506 by the centering processing unit 507 and then the preprocessing by the preprocessing unit 101 is performed. . In this case, the shape image after the centering process corresponds to the “shape image created by the phase continuation processing means” in the present invention. The centering process is not essential in the present invention. In particular, in the second embodiment, the value of each line of the shape image is integrated and averaged in the longitudinal direction of the measurement object 2 to create the reference shape image. It is desirable to perform a centering process.

  The image processing apparatus 50 to which the present invention is applied is realized by a computer system including a CPU, a ROM, a RAM, and the like, and the functions or processing of the image processing apparatus 50 are realized by the CPU executing a computer program. Note that the image processing apparatus 50 does not have to be a single apparatus, and may include a plurality of devices.

  In addition, a software program for realizing the functions of the above-described embodiments for an apparatus or a computer in the system connected to the various devices so that the various devices are operated to realize the functions of the above-described embodiments. What was implemented by supplying the code and operating the various devices in accordance with a program stored in a computer (CPU or MPU) of the system or apparatus is also included in the scope of the present invention. In this case, the program code of the software itself realizes the functions of the above-described embodiment, and the program code itself and means for supplying the program code to the computer, for example, a record storing the program code The medium constitutes the present invention. As a recording medium for storing the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

It is a schematic block diagram of a comparative surface defect inspection system. It is a figure which shows the example of the cross-sectional shape of a measuring object. It is a figure for demonstrating the structure and operation | movement of a delay integration type camera. It is a figure which shows an example of a fringe image. It is a figure for demonstrating the shift | offset | difference of the stripe in a stripe image. It is a figure which shows an example of a fringe image. It is a figure for demonstrating the relationship between the shift | offset | difference of the phase in slice fringe image data, and the depth of a measuring object. It is a schematic block diagram of an image processing apparatus. It is a figure which shows the photograph of the image in each step, (a) is a figure which shows the photograph of a phase image, (b) is a figure which shows the photograph of a shape image, (c) is a figure which shows the photograph of a depth image. is there. It is a characteristic diagram showing the phase characteristic in each stage, (a) is a characteristic figure showing the phase characteristic in a phase image, (b) is a characteristic figure showing the phase characteristic in a shape image, (c) is a phase in a depth image. It is a characteristic view showing a characteristic. It is a figure which shows the photograph of an amplitude image. It is a figure for demonstrating the outline | summary of a centering process, (a) is a schematic diagram which shows a mode when a measuring object meanders, (b) is a schematic diagram which shows a mode when a measuring object vibrates, ( c) is a schematic diagram showing an image after the centering processing. It is a figure which shows the photograph of the image used for surface defect discrimination | determination, (a) is a figure which shows the photograph of a binarized image, (b) is a figure which shows the photograph of a depth image, (c) is a photograph of a brightness | luminance image. FIG. It is a flowchart for demonstrating the processing operation of the comparative image processing apparatus. It is a schematic block diagram of the surface defect inspection system which concerns on embodiment. 6 is a flowchart for explaining a processing operation of the image processing apparatus according to the embodiment. It is a figure for demonstrating the outline | summary of a fitting process, (a) is a figure which shows a mode that a fitting curve is dragged to the surface defect in contrast, (b) is fitting to a surface defect by masking in embodiment. It is a figure which shows a mode that a curve is not dragged. It is a figure for demonstrating the outline | summary of the fitting process in a primary shape correction process, and a difference calculation. It is a figure for demonstrating the outline | summary of a binarization process and an expansion process, (a) is a figure which shows a binarized image, (b) is a figure which shows the image after an expansion process. It is a figure for demonstrating the outline | summary of an expansion process, (a) is a figure which shows the example which expands only in the width direction, (b) is a figure which shows the example which expands in a two-dimensional direction. It is a figure for demonstrating an example of the result of the surface defect detection in contrast, (a) is a figure which shows the photograph of a depth image, (b) is a characteristic view. It is a figure for demonstrating an example of the result of the surface defect detection in embodiment, (a) is a figure which shows the photograph of a depth image, (b) is a characteristic view. It is a figure showing the actual value of an examiner, the comparison and the result of the surface defect inspection by this embodiment. It is a figure for demonstrating the outline | summary of the preparation process and difference calculation of the reference | standard shape image in a primary shape correction process. It is a figure for demonstrating the outline | summary of a fitting process, (a) is a figure which shows a mode which should be fitted originally, (b) is a figure which shows a mode that the fitting curve is dragged by the surface defect.

Explanation of symbols

2 Measurement object 10 Laser device 20 Rod lens 30 Delay integration type camera 31 Lens 40 Timing signal generator 50 Image processing device 60 Display device 71 Fitting curve 72 Shape image after centering processing 73 Rebound 74 Threshold value for binarization 75 Reference shape image 101 Pre-processing unit 501 A / D conversion unit 502 Pre-filter unit 503 Quadrature sine wave generation unit 504a, 504b Low-pass filter unit 505 Phase calculation unit 506 Phase continuation processing unit 507 Centering processing unit 508 Shape correction processing unit 509 Defect Detection processing unit 510 Amplitude calculation unit 511 Edge position detection unit

Claims (6)

In the cross-sectional shape, the measurement object is a long material having a curve in which both ends become lower or higher,
Using irradiation means for irradiating periodically modulated linear laser light in the width direction of the surface of the measurement object, and delay integration type imaging means for imaging reflected light from the measurement object, While continuously shifting the irradiation position of the linear laser beam on the measurement object, the reflected light from the measurement object is imaged by the imaging means to output a light cut image, and surface defects of the measurement object are detected. A surface defect inspection system for detecting,
Phase calculating means for calculating a phase shift at each position of the fringe image obtained by sequentially arranging the light sectioned images obtained by the imaging means;
Based on the phase shift obtained by the phase calculation means, a position where the phase shift is discontinuous is detected, and the phase shift at the detected position is connected to make the phase shift continuous and the measurement is performed. Phase continuation processing means for creating a shape image representing the surface shape of the object;
Pre-processing means for extracting irregularities on the shape image created by the phase continuation processing means;
A fitting process for obtaining a fitting curve is performed on the shape image obtained by removing the irregularities extracted by the pre-processing means from the shape image created by the phase continuation processing means, and the fitting curve and the phase continuation There rows difference operation between shape image generated in the process unit, the data obtained in the difference calculation, on average by a predetermined length in the longitudinal direction of the measurement target, a low-pass filter to the averaged value and treatment was subjected to a value, and again the line cormorant shape correcting means for difference calculation between the obtained data in the difference calculation,
A surface defect inspection system comprising: defect detection processing means for detecting a surface defect of the measurement object based on data obtained by difference calculation of the shape correction processing means. The pre-processing means includes
A primary shape correction for performing a fitting process for obtaining a fitting curve on the shape image created by the phase continuation processing means and performing a difference calculation between the fitting curve and the shape image created by the phase continuation processing means Processing means;
Binarization means for binarizing data obtained by the difference calculation of the primary shape correction processing means;
A concavo-convex part extracting means for extracting the concavo-convex part from the binarized image binarized by the binarization means;
The mask processing means for generating mask information for removing the uneven portion extracted by the uneven portion extracting means from the shape image created by the phase continuation processing means. Surface defect inspection system. The pre-processing means includes
The shape image created by the phase continuation processing means is integrated and averaged in the longitudinal direction of the measurement object to create a reference shape image, and the reference shape image and the shape image created by the phase continuation processing means Primary shape correction processing means for performing a difference calculation with
Binarization means for binarizing data obtained by the difference calculation of the primary shape correction processing means;
A concavo-convex part extracting means for extracting the concavo-convex part from the binarized image binarized by the binarization means;
The mask processing means for generating mask information for removing the uneven portion extracted by the uneven portion extracting means from the shape image created by the phase continuation processing means. Surface defect inspection system. Expansion processing means for performing expansion processing for increasing the uneven portions extracted by the uneven portion extracting means at least at a predetermined rate in the width direction of the measurement object;
4. The mask processing unit according to claim 2 or 3, wherein the mask processing unit generates mask information for removing the uneven portion expanded by the expansion processing unit from the shape image generated by the phase continuation processing unit. The surface defect inspection system described. In the cross-sectional shape, the measurement object is a long material having a curve in which both ends become lower or higher,
Using irradiation means for irradiating periodically modulated linear laser light in the width direction of the surface of the measurement object, and delay integration type imaging means for imaging reflected light from the measurement object, While continuously shifting the irradiation position of the linear laser beam on the measurement object, the reflected light from the measurement object is imaged by the imaging means to output a light cut image, and surface defects of the measurement object are detected. A surface defect inspection method to detect,
A phase calculation procedure for calculating a phase shift at each position of the fringe image obtained by sequentially arranging the light section images obtained by the imaging means;
Based on the phase shift obtained in the phase calculation procedure, a position where the phase shift is discontinuous is detected, and the phase shift at the detected position is connected to make the phase shift continuous, and the measurement is performed. A phase continuation processing procedure for creating a shape image representing the surface shape of the object;
A pre-processing procedure for extracting irregularities on the shape image created by the phase continuation processing procedure;
A fitting process for obtaining a fitting curve is performed on the shape image obtained by removing the uneven portion extracted by the pre-processing procedure from the shape image created by the phase continuation processing procedure, and the fitting curve and the phase continuation are performed. There rows difference calculation between the shape image generated by the processing procedure, the data obtained in the difference calculation, on average by a predetermined length in the longitudinal direction of the measurement target, a low-pass filter to the averaged value and treatment was subjected to a value, and again the line cormorant shape correction procedure the difference calculation between data obtained by the difference calculation,
A surface defect inspection method comprising: a defect detection processing procedure for detecting a surface defect of the measurement object based on data obtained by difference calculation of the shape correction processing procedure. In the cross-sectional shape, the measurement object is a long material having a curve in which both ends become lower or higher,
Using irradiation means for irradiating periodically modulated linear laser light in the width direction of the surface of the measurement object, and delay integration type imaging means for imaging reflected light from the measurement object, While continuously shifting the irradiation position of the linear laser beam on the measurement object, the reflected light from the measurement object is imaged by the imaging means to output a light cut image, and surface defects of the measurement object are detected. A program for detecting,
A phase calculation process for calculating a phase shift at each position of the fringe image obtained by sequentially arranging the light sectioned images obtained by the imaging means;
Based on the phase shift obtained in the phase calculation process, the position where the phase shift is discontinuous is detected, and the phase shift at the detected position is connected to make the phase shift continuous, and the measurement is performed. A phase continuation process for creating a shape image representing the surface shape of the object;
A pre-processing for extracting irregularities on the shape image created by the phase continuation processing;
A fitting process for obtaining a fitting curve is performed on a shape image obtained by removing the uneven portions extracted by the pre-processing from the shape image created by the phase continuation process, and the fitting curve and the phase continuation process There rows difference operation between created shape image, the data obtained in the difference calculation, on average only longitudinally for a predetermined length of the measurement target, facilities low-pass filter processing on the average value a value obtained by a re-row power sale shape correction process the difference calculation between the obtained data in the difference calculation,
A program for causing a computer to execute a defect detection process for detecting a surface defect of the measurement object based on data obtained by the difference calculation of the shape correction process.

Images