JP5857675B2 - Surface defect inspection apparatus and surface defect inspection method - Google Patents

Description

  The present invention relates to a technique for detecting a surface defect, and mainly relates to a surface defect inspection apparatus and a surface defect inspection method for reliably detecting a defect or a flaw defect on a paint surface.

  On the painted surface of resin molded products such as mobile phones, a base layer mainly containing pigments, etc., and a clear layer on the base layer for glossing are provided and painted in two layers. . Because the clear surface is transparent, it is difficult to detect irregularities on the painted surface with normal illumination. Instead, a method using a black and white striped illumination pattern in which bright and dark areas are arranged alternately is used. Are known. In this method, an illumination pattern is irradiated on a coating surface to be inspected, and a defect is found in a dark part of the illumination pattern reflected from the surface. In other words, the irradiated light is almost regularly reflected on the paint surface, but if there is a defect, the light emitted from the bright part is reflected by the inclined part of the defect, and the reflected light whose reflection angle has changed in the inclined part appears in the dark part. Defects can be revealed.

  Although a method using a bright and dark illumination pattern can visually identify a defect, there is an oversight or the like due to visual observation. Therefore, a method for automatically detecting a defect has been proposed. In this method, the surface irradiated with the illumination pattern is imaged by a CCD (Charge Coupled Device Image Sensor) camera or the like, and the luminance of each pixel of the obtained image is different from that of each pixel of the image having no defect. When there is a part, it is determined as a defect. When the illumination pattern is in the form of black and white stripes, it is an effective method for detecting defects in the case of flaws caused by dust and the like, but in the case of defects with directionality such as scratches, Depending on the direction, it may be difficult to detect. For this reason, a method has been proposed that uses an illumination pattern in which the light and dark are in a check shape or a jagged shape.

  In addition, since the illumination pattern is a binary black and white pattern, when the defect is in the dark part pattern, light does not reach the defective part, and the illumination pattern area has a high detection sensitivity area and a low detection area. There is a problem of non-uniformity in detection sensitivity.

Japanese Patent Laid-Open No. 04-231853 Japanese National Patent Publication No. 10-509238 JP 2000-018932 A

  As described above, a method for illuminating the inspection surface by illuminating a bright and dark illumination pattern on the inspection surface is used for defect inspection of the coating surface. A striped light / dark pattern is generally used as the illumination pattern, but it may be difficult to detect a directional defect depending on the direction of the defect. To improve the detection of directional defects, it has also been proposed to use lighting patterns with check and jagged light and darkness, but it is not sufficient to deal with defects in all directions, but it is sufficient. There wasn't. Further, since these illumination patterns use binary black and white patterns, there is a problem of non-uniform detection sensitivity in the illumination pattern area.

  In view of the above problems, an object of the present invention is to provide a surface defect inspection apparatus and a surface defect inspection method that reliably detect even a directional defect that has no non-uniformity in detection sensitivity.

  According to an aspect of the invention, an illumination unit that irradiates a reference surface and an inspection surface with an illumination pattern whose hue periodically changes in at least two directions on a plane, the reference surface irradiated with the illumination pattern, and the Provided is a surface defect inspection apparatus comprising: an imaging unit that images an inspection surface; and a defect detection unit that detects the presence or absence of a defect on the inspection surface based on a difference in hue between both images captured by the imaging unit. it can.

  According to another aspect of the invention, the saturation is modulated in a sine wave shape in the first direction of the plane, and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction. An illumination unit that irradiates the illumination pattern that has been changed by moving the illumination pattern a plurality of times on the reference surface and the inspection surface, an imaging unit that captures the reference surface and the inspection surface each time the movement is performed, and imaging by the imaging unit A first evaluation value for a saturation change is obtained from the plurality of images of the reference surface, and a second evaluation value for a saturation change is obtained from the plurality of images of the inspection surface, and the first evaluation value and the It is possible to provide a surface defect inspection apparatus having a defect detection unit that detects the presence or absence of defects on the inspection surface based on the difference between the second evaluation values.

  According to another aspect of the invention, the hue and saturation are modulated in a sine wave shape with a predetermined phase difference in the first direction of the plane, and a second direction perpendicular to the first direction is modulated. An illumination unit that illuminates the reference surface and the inspection surface by moving an illumination pattern that periodically changes the phase of the sine wave in the direction, and the reference surface and the inspection surface that are irradiated with the illumination pattern, A first evaluation value for a hue change is obtained from a plurality of images of the reference plane imaged by the imaging section, and a second change for a hue change from the plurality of images on the inspection surface. And a defect detection unit that detects a defect on the inspection surface based on a difference between the first evaluation value and the second evaluation value.

  According to another aspect of the invention, the hue is modulated in a sine wave shape in a first direction of a plane, and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction. And moving the illumination pattern whose saturation is modulated into the sine wave to illuminate the reference surface and the inspection surface, and moving the reference surface and the inspection surface irradiated with the illumination pattern. A first evaluation value for a hue change from a plurality of images of the reference plane imaged by the imaging unit, and a second evaluation for a hue change from the plurality of images of the inspection surface It is possible to provide a surface defect inspection apparatus having a defect detection unit that obtains a value and detects the presence or absence of a defect on the inspection surface based on a difference between the first evaluation value and the second evaluation value.

  According to another aspect of the invention, an illumination procedure for irradiating a reference surface and an inspection surface with an illumination pattern whose hue periodically changes in at least two directions on a plane, and the reference surface irradiated with the illumination pattern And a defect detection procedure for detecting the presence or absence of a defect on the inspection surface based on a difference in hue between both images captured by the imaging unit. Can provide.

  According to the present invention, since the illumination pattern whose hue or saturation periodically changes in at least two directions on the plane is used, the nonuniformity of the detection sensitivity is eliminated and even a directional defect is eliminated. A surface defect inspection apparatus and a surface defect inspection method that can be reliably detected can be provided.

It is a figure which shows the example of the surface defect inspection method. It is a figure which shows the captured image example. It is a figure which shows the example of a defect detection sensitivity of an illumination pattern. It is a figure which shows the example of a captured image by the illumination pattern of a sine wave luminance modulation. It is a figure which shows the luminance change and phase shift of the image of a defective part. It is a figure which shows the example of the picked-up image of the defect which has directionality. It is a figure which shows a check pattern and a jagged pattern example. It is a figure which shows the structural example (Example 1) of the surface defect inspection apparatus of this invention. It is a figure which shows the example (Example 1) of the illumination pattern which changed the hue to all directions. It is a figure which shows the omnidirectional pattern example 1 (Example 1). It is a figure which shows the example of hue modulation (Example 1) of the example 1 of an omnidirectional pattern. It is a figure which shows the saturation modulation example (Example 1) of the example 1 of an omnidirectional pattern. It is a figure which shows the example 2 of omnidirectional pattern (Example 1). It is a figure which shows the example of a flow (Example 1) of a defect inspection. It is a figure which shows the structural example (Example 2) of the surface defect inspection apparatus of this invention. It is a figure which shows the example 1 (Example 2) of the hue change in a test surface position. It is a figure which shows the example 2 (Example 2) of the hue change in a test | inspection surface position. It is a figure which shows the example of a flow (Example 2) of a defect inspection. It is a figure which shows the lighting pattern example 1 (Example 3) which modulated the hue and the saturation in the same direction. It is a figure which shows the lighting pattern example 2 (Example 3) which modulated the hue and the saturation in the same direction. It is a figure which shows the example 1 (Example 4) of the illumination pattern which modulated the hue and the saturation orthogonally. It is a figure which shows the example 2 (Example 4) of the illumination pattern which modulated the hue and the saturation orthogonally.

  Before describing an embodiment of the present invention, a surface defect inspection method and various illumination patterns will be described with reference to FIGS. Here, the inspection target is the surface of a coating film painted on a mobile phone or the like, and a defect on the surface having no directionality or a flaw defect having directionality will be described as an example.

  FIG. 1 is a diagram showing an example of a surface defect inspection method. An inspection surface 10 to be inspected is composed of a coating film including a base layer 11 and a clear layer 12 on the base layer 11. A foreign substance 20 such as dust adheres to the base layer 11, and the clear layer 12 painted on the foreign substance 20 shows a state where there is a convex defect 21 having a slope (in this case, a flaw defect).

  The illumination unit 30 that illuminates the inspection surface 10 includes an illumination pattern that includes a bright part 31 and a dark part 32. Here, the bright portions 31 and the dark portions 32 are alternately arranged in a stripe shape.

  The imaging unit 40 images the inspection surface 10 illuminated by the illumination unit 30. Since the light emitted from the illumination unit 30 is regularly reflected on the inspection surface 10, the imaging unit 40 images the illumination pattern reflected on the inspection surface 10. When the light emitted from the illumination unit 30 strikes the defect 21 on the inspection surface 10, the imaging unit 40 is at an angle different from the angle of the light reflected by the inclined surface of the defective portion and reflected by the inspection surface 10 without the defect. Is incident on. A light 41 indicated by a solid line in FIG. 1 indicates light that is specularly reflected on the inspection surface 10 and is incident on the imaging unit 40. The incident light is shown.

  Next, the captured image 50 captured by the imaging unit 40 will be described with reference to FIG. The captured image 50 shown in FIG. 2 is an image of the illumination pattern reflected by the inspection surface 10 shown in FIG. 1, and the stripe-like shape of the bright part 51 and the dark part 52 corresponding to the bright part 31 and the dark part 32 of the illumination part 30. Is imaged. The defect 21 is a portion shown as a white spot in the stripe of the dark portion 52. In this way, the defect can be revealed by irradiating a bright and dark illumination pattern.

  However, the defect 21 cannot be manifested at any position of the bright part 31 and the dark part 32 irradiated with the illumination pattern, and the detection sensitivity differs depending on the position. FIG. 3A shows the luminance distribution of the region of bright part 31 -dark part 32 -bright part 31 on the inspection surface irradiated with the stripe illumination pattern. Since the illumination pattern is a black and white binary stripe pattern, the luminance distribution sharply decreases from the bright portion 31 to the dark portion 32 and rapidly increases from the dark portion 32 to the bright portion 31 as shown in FIG. Will be shown. FIG. 3B shows the defect detection sensitivity at a position corresponding to the luminance distribution shown in FIG. There are two areas around the position where the luminance is low to the position where the luminance is low, that is, two peak areas at both ends of the dark part 32, and there are peak areas with high detection sensitivity. If there is a defect 21 in this region, the light 21 diffused directly from the bright part 31 of the illumination unit 30 or the diffused light is reflected on the inclined surface of the defect 21 and forms an image in the dark part 52 of the imaging unit 40 so that the defect 21 can be identified. It becomes like this. Compared with the end of the dark part 32 on the inspection surface 10, the amount of light diffusing from the bright part 31 is reduced in the central part of the dark part 32, so that the detection sensitivity is low. Therefore, the defect is detected with the highest sensitivity when the defect is located at the peak region. As described above, the black-and-white binary illumination pattern includes a region having a high defect sensitivity and a region having a low defect sensitivity depending on the region. For this reason, in the defect inspection, it is necessary to move the illumination pattern or the inspection surface little by little so that the entire region of the surface to be inspected passes through this peak region.

  In order to improve the nonuniformity of the defect sensitivity, the present applicant has proposed a method using a striped illumination pattern whose luminance changes in a sine wave shape (Japanese Patent Application No. 2011-79528). This method uses the phase shift method, which detects the defect by moving the above illumination pattern and imaging it, and comparing the amplitude or phase of the luminance displacement between the image without defect and the image with defect. is there. FIG. 4 shows an image captured using a striped illumination pattern in which the luminance changes in a sine wave shape. In FIGS. 4A to 4D, the illumination pattern is moved and irradiated, and the phase of the illumination pattern irradiated on the inspection surface is displaced every π / 2 from 0 to 3π / 2 of one cycle of the sine wave. It is an image. The circular image changing from white to black shown in the images of FIGS. 4A to 4D is a defect, and the luminance of the defect image changes with the phase change caused by the movement of the illumination pattern. It is shown that. The change in luminance of the image of the defective part and the change in luminance of the image of the part having no defect at the same position as the defect are shown in FIG. 5, and there is a difference between the amplitude and phase of the luminance change. That is, the amplitude of the luminance change of the image of the defective portion is small and the phase is slightly advanced as compared with the image of the portion having no defect. A defect is detected based on the difference in the amplitude or the phase of the luminance change.

  Incidentally, the defect in the image shown in FIG. 2 is a defect having no directionality, but in the case of a defect having a directionality such as a scratch, the direction of the scratch becomes a problem. FIG. 6A shows an example in which the direction of the scratch defect 22 is parallel to the direction of the vertical stripe of the illumination pattern (that is, the vertical scratch defect), and the upper figure shows the inspection surface 10 having the scratch defect 22. The lower figure is an image obtained by imaging the inspection surface 10. Since the inclined surface due to the flaw defect 22 is formed along the same vertical direction as the stripe, the light reflected by the inclined surface can be clearly identified as a white line in the dark portion 52 of the captured image. On the other hand, FIG. 6B shows an example in which the direction of the flaw defect 23 is perpendicular to the direction of the light and dark stripes (that is, the flaw defect in the lateral direction), and the inclined surface is formed in the direction perpendicular to the stripe. ing. For this reason, since the direction of the light reflected by the inclined surface is reflected in the same vertical direction as the stripe, it is difficult to identify a defect in the dark portion 52 of the captured image. That is, when a striped illumination pattern is used, it may not be detected depending on the direction of the defect.

  In order to improve this problem, a check pattern shown in FIG. 7A and a jagged pattern shown in FIG. 7B have been proposed. For example, in the check pattern, since the four sides of the dark part are in contact with the bright part, it is possible to detect a flaw defect in the vertical direction and the horizontal direction. In the jagged pattern, a flaw defect having a 45 ° or 135 ° direction can be detected. These are improved compared to those limited to one direction of the stripe pattern, but do not correspond to all directions. Further, the striped illumination pattern whose luminance is changed to a sine wave shape shown in FIG. 4 is not considered for a directional defect.

  Next, embodiments of the present invention will be described in Examples 1 to 4.

Example 1
In the first embodiment, an illumination pattern whose hue or saturation is always displaced with respect to all directions of a plane is used as an illumination pattern irradiated on the inspection surface.

  First, the structure of the surface defect inspection apparatus in Example 1 is demonstrated using FIG. The surface defect inspection apparatus 100 is roughly composed of a control unit 200, an illumination unit 300, and an imaging unit 400. The elements and functions constituting each of these parts will be described.

  The control unit 200 includes a CPU 210 that controls the entire surface defect inspection apparatus 100, a RAM (Random Access Memory) 220 that executes and executes an inspection program on a memory, a keyboard (KB) 271 that receives input from an operator, and a process A display (DISP) 272 for displaying results and the like, an input / output interface (IO / IF) 270 for exchanging data, an image data storage unit 230 for storing image data captured from the imaging unit 400 described later, and an inspection program are stored. The inspection program storage unit 231, the imaging command control command is transmitted to the imaging unit 400, the imaging control unit (imaging CNT) 240 that captures image data captured from the imaging unit 400, and the illumination unit 300 is turned on-off. Lighting control unit that sends out control commands ( Equipped with a bright CNT) 250.

  The illumination unit 300 is installed in parallel above the inspection surface 10 placed horizontally, includes an illumination pattern panel 310, and irradiates the inspection surface 10 with an illumination pattern. The illumination pattern panel 310 is obtained by pasting a film on which an illumination pattern is drawn on a surface-emission illumination panel.For example, the illumination pattern displayed on the liquid crystal display element transmits light from a light source disposed on the back surface. You may make it project on the test | inspection surface 10 through a lens.

  The imaging unit 400 is a camera using, for example, a CCD image sensor as an imaging element, and performs color imaging of the inspection surface 10 by a control command instructed from the imaging CNT 240. The captured image data is stored in the image data storage unit 230 via the imaging CNT 240. The installation position of the imaging unit 400 is arranged at a position where the illumination pattern irradiated from the illumination unit 300 is regularly reflected by the light receiving unit of the imaging unit 400 via the inspection surface 10 and imaged.

  Next, an illumination pattern will be described. The illumination pattern shown in FIG. 9A is the illumination pattern used in the first embodiment, and here is a pattern created so that the hue is displaced with respect to all directions on the plane. In this example, the hue is modulated in a sine wave shape in the X direction in the figure, and the phase of the sine wave is periodically changed in the Y direction. In the figure, the halftone dots are red (R), the diagonal lines are green (G), and the horizontal lines are blue (B).

  The arrows of (1) to (4) shown in FIG. 9A indicate the directions on the plane of the illumination pattern, and these four directions are selected as representatives in order to explain that the hue changes in all directions. It is. FIG. 9B is a diagram showing changes in hue corresponding to these four directions. As shown in FIG. 9B, the hue always changes in any of the directions (1) to (4). The change in hue means that the hue at a position where the phase is shifted by π from the position at which the hue ring is located (angle) changes to the maximum.

  Although FIG. 9 shows an example in which the hue changes in all directions, the saturation may similarly be changed in all directions with the hue kept constant.

  The illumination pattern will be described in more detail. FIG. 10 is an example of a pattern in which the hue changes in all directions in the same way as the illumination pattern shown in FIG. 9A, and the hue is expressed by Expressions (1) and (2).

  Here, x and y indicate the x and y coordinates in the plane of the illumination pattern, t is a variable (phase), A is the amplitude of hue modulation, and constants a to c are the periods in the illumination pattern shown in FIG. Is shown. Further, θ (y) represents a phase shift and is a function that changes periodically according to the displacement of y. a to c are appropriately determined according to the size of the defect to be inspected and the size of the undulations.

  FIG. 11 shows a state in which the intensity and hue of RGB change in the illumination pattern that is hue-modulated by the equations (1) and (2). FIG. 11A shows the pattern shape of the illumination pattern, FIG. 11B shows a case where the degree of hue modulation is expressed by luminance, and FIG. 11C shows the arrow shown in FIG. An intensity change of RGB at the position in the direction is shown, and FIG. 11D shows a change in hue.

  As shown in FIG. 11 (c), the intensity displacement of each RGB is not sinusoidal, but the hue is converted into a hue by HSV (Hue, Saturation, Value) conversion as shown in FIG. 11 (d). The displacement of is sinusoidal.

  Using the expressions (1) and (2), the saturation can be modulated in all directions of the illumination pattern. FIG. 12 is a diagram showing this example. FIG. 12A shows the pattern shape of the illumination pattern when the hue is red, and FIG. 12B shows the case where the degree of saturation modulation is expressed by luminance. 12C shows the intensity change of RGB at the position in the direction of the arrow shown in FIG. 12A, and FIG. 12D was obtained by HSV conversion of RGB shown in FIG. 12C. A change in saturation indicates a sine wave displacement.

  An illumination pattern whose hue or saturation changes in all directions can also be created using equation (1) and, for example, equation (3). The shape of the illumination pattern and the constants a to c are shown in FIG.

  Next, the processing flow of the surface defect inspection apparatus 100 will be described using the processing flow example of FIG. Here, an example using an illumination pattern whose hue changes as shown in FIG. 10 will be described.

  In FIG. 14, first, an illumination pattern is irradiated onto the surface of a coating film (herein referred to as a reference plane) that is known to have no defects on the surface in advance, and color imaging is performed by the imaging unit 400. The captured image data is stored in the image data storage unit 230. Subsequently, an illumination pattern is similarly applied to the surface to be inspected (herein referred to as an inspection surface), and color imaging is performed. The captured image data is stored in the image data storage unit 230. Thus, two image data of the reference surface and the inspection surface are obtained (S1-S4. Hereinafter, step 1 is expressed as S1).

  The image data of the reference surface and the inspection surface is extracted from the image data storage unit 230, and the hues of the pixels at the same position in both images are compared. If the difference in hue is larger than a predetermined value, it is regarded as a pixel (herein referred to as a defective pixel) in which a defective portion on the surface is imaged. All pixels of the image data are compared, and defective pixels are extracted (S5).

  The size of the defect composed of defective pixels is obtained. If the defect size is smaller than a predetermined value, it is regarded as acceptable, and a display of acceptance is displayed on the display 272. If the defect size is larger than the predetermined value, the defect is regarded as defective. A pass is displayed (S6-S9).

  Thus, the defect can be detected. Since the hue of the illumination pattern changes in all directions, the hue of the light reflected by the inclined portion of the flawed defect appears in another hue, and the defect can be detected. The processing flow of FIG. 14 uses the illumination pattern of FIG. 10 in which the hue is modulated, but the same flow can be detected using the illumination pattern of FIG. 12 in which the saturation is modulated. Moreover, even if the illumination pattern shown in FIG. 13 is used, the hue or saturation changes in all directions, and a defect can be detected.

(Example 2)
In the first embodiment, an illumination pattern in which the hue or saturation is changed in all directions is used, the reference surface and the inspection surface are imaged, and a defect image is extracted from the difference in hue or saturation between the two images. It was. In the second embodiment, an illumination pattern in which the hue or saturation is changed in all directions is used in the same manner as in the first embodiment. However, the illumination pattern is moved and irradiated, and the amplitude of the hue or saturation hue is extracted for extracting defective pixels. This is an example of using the difference or phase difference.

  Initially, the structural example of the surface defect inspection apparatus of Example 2 is demonstrated using FIG. The surface defect inspection apparatus 101 includes a control unit 201, an illumination unit 301, and an imaging unit 400. Since the surface defect inspection apparatus 101 is partially different in configuration from the surface defect inspection apparatus 100 described in the first embodiment, this different part will be described.

  In order to move the illumination pattern, which is a feature of the second embodiment, the illumination unit 301 includes a moving mechanism 320. The moving mechanism 320 supports a light source (not shown) and the illumination pattern panel 310 and moves according to a command from the control unit 201. The illumination unit 301 is installed in parallel above the inspection surface 10 placed horizontally, and the movement of the illumination pattern 310 is moved in accordance with the command while maintaining the parallel movement (the arrow in FIG. 15). Indicates the direction of movement). In addition, the control unit 201 includes a movement control unit (moving CNT) 260 for instructing the lighting unit 301 to move. In addition, an inspection program storage unit 232 storing an inspection program is provided (the inspection program here is different from the inspection program stored in the inspection program storage unit 231 described in the first embodiment, and movement control and movement of the illumination pattern). The image processing obtained by the above is performed). The configurations other than those described above are the same as those of the surface defect inspection apparatus 100. In addition, the code | symbol attached | subjected to the surface defect inspection apparatus 101 shown in FIG.

  Next, in the second embodiment, the illumination pattern irradiated on the inspection surface 10 is moved, and the movement amount and the hue change when moved are described.

  FIG. 16 is a diagram illustrating an illumination pattern on the inspection surface 10 captured by the imaging unit 400 by movement. The illumination pattern is moved three times for each ¼ distance of the period a of the hue-modulated sine wave. Due to this movement, the illumination surface is irradiated with the illumination pattern shifted by the movement amount. This means that the illumination patterns in which the phase t of the sine wave in hue modulation is displaced by π / 2 (that is, four types of illumination patterns in which the phase t is changed by π / 2 from 0 to 3π / 2) are sequentially switched. This is equivalent to that irradiated on the inspection surface 10. A, B, and C on the inspection surface in FIG. 16A indicate positions where the hues are R (red), G (green), and B (blue) at t = 0, and FIG. FIG. 16D shows a state where the hues at the respective positions A to C are changed when the illumination pattern is moved three times. If the illumination pattern is moved continuously, the hue changes as shown in FIG. 17, although the initial hues (t = 0) at the respective positions A to C are different. The hues at positions A to C change at the same period.

  The hue change at positions A to C shown in FIG. 17 shows a state of specular reflection that is not a defective portion. If there is a defect at this position, the amplitude and phase of the hue change are shown in FIG. The amplitude of the hue change is smaller than that indicated by 17, and the phase is shifted. Similar to the amplitude and phase shift of the luminance change described with reference to FIG. 5, the reflected light is changed at the inclined portion of the defect and the reflected light is imaged at a position different from the position where it should be regularly reflected ( That is, a phase shift occurs), and the amplitude of the hue change at that position becomes small due to diffusion due to reflection. In the second embodiment, defective pixels are extracted by comparing with the amplitude or phase of hue change in the case of regular reflection. By using a method of comparing the amplitude or phase of the hue change (that is, the phase shift method), it is possible to detect the defective pixel with higher detection accuracy in the first embodiment.

  Next, the process flow of the surface defect inspection will be described using the process flow example of FIG. In this case, the illumination pattern shown in FIG. 10 is moved three times in the X direction by a distance 1/4 of the period a for modulating the hue of the sine wave, and an image is obtained by imaging the inspection surface 10 irradiated with the illumination pattern. (That is, the inspection surface 10 is imaged by irradiating the illumination pattern with the phase t advanced by π / 2). Thereby, an image captured four times in total, that is, the initial position of the illumination pattern and the position moved three times, is obtained. The number of times of imaging should be at least 3 times (the hue is modulated by a sine wave, so the constant of the sine wave can be specified by obtaining the value of at least 3 points), and if the number of times of imaging is increased, the accuracy will be Rise. Prior to this inspection, the image data storage unit 230 stores image data (referred to as reference image data) captured in a processing flow described below on a reference surface having no defect. To do. Note that the processing flow of FIG. 18 is a flow processed by the CPU 210 of the surface defect inspection apparatus 101.

  First, the CPU 210 initializes a counter n that counts the number of photographing (n = 0) (S11).

  Since the counter n is incremented (n = n + 1) and n = 1, a control command is sent to the moving mechanism 320 to move the illumination pattern from the standby position to the initial position where the first imaging is performed. (The moving mechanism is assumed to be in the standby position when n = 0). When the position of the illumination pattern is determined, a control command is sent to the imaging unit 400, and the imaging surface 400 images the inspection surface 10. The captured image data is stored in the image data storage unit 230 (S12-S14, S16, S17).

  Since the counter n is smaller than “4” at which imaging is performed (n = 1 at this time), the process returns to S12 to count up and move the illumination pattern a distance of a / 4 in the X direction in FIG. By this movement, the inspection surface 10 is irradiated with an illumination pattern in which the phase t of the sine wave is displaced in the π / 2X direction from the illumination pattern captured for the first time. Imaging is performed after movement, and the obtained image data is stored. This is repeated from S12 to S18 until n = 4. As a result, the imaging unit 400 performs imaging four times, and the image data storage unit 230 stores four image data (referred to as inspection image data) (S12, S13, S15-S18). ).

When the four imaging operations are completed, detection pixels are extracted based on the reference image data captured in advance from the acquired four inspection image data. The defective pixel can be extracted as follows, for example. First, an evaluation value δ of a pixel at each position of the image is obtained from the four inspection image data by the equation (4). On the other hand, also for the reference image data, the evaluation value δ ₀ is obtained using the calculation formula (4). Next, a difference (δ−δ ₀ ) between the obtained δ and δ ₀ is calculated, and when the calculated difference exceeds a predetermined value (threshold), it is regarded as a defective pixel.

In Equation (4), I _{1 to} I ₄ are hue values obtained by HSV conversion of the RGB values of each pixel of the x coordinate and y coordinate when the phase t is displaced to 0 to 3π / 2 (S19).

  Based on the number of extracted defective pixels, the size (size) of the defect is obtained. If the size of the defect is smaller than a predetermined size, the result is regarded as acceptable, and an indication of acceptance is displayed on the display 272. If it is larger, it is regarded as a defect and a failure is displayed (S20-S23).

  In the above description, the processing flow has been described using an example in which an illumination pattern that changes the hue in FIG. 10 is used. However, the same processing flow is obtained even if the illumination pattern in which the saturation is changed in FIG. Further, the difference in hue amplitude is used for extracting defective pixels, but a phase difference may be used instead.

  In the above flow, the description of the on-off control for the light source of the illumination unit 301 in the illumination control unit 250 of FIG. 15 is omitted. However, for example, the light source is turned on during S12 to S18 repeated four times, and the others May be controlled to be off. In the second embodiment, the illumination mechanism is moved by the movement mechanism 320 to shift the phase. However, the illumination pattern may be switched without using the movement mechanism. In this case, an illumination pattern whose phase is displaced by t = 2πk / n (k = 0, 1, 2,...) Is created, and the illumination pattern is switched.

(Example 3)
In the second embodiment, an illumination pattern in which the hue is changed in all directions is used. However, in the third embodiment, the hue and saturation are given a phase difference, and these are changed in all directions to use the illumination pattern. It is an example. The modulation of the hue I _hue (x, y, t) and the saturation I _sat (x, y, t) at this time is expressed by, for example, the expressions (5) and (6), and θ (y) It is shown by the formula (2).

  As shown by the equations (5) and (6), the saturation modulation in the X direction is advanced by π / 2 with respect to the hue modulation, and the illumination pattern is as shown in FIG. become. The RGB intensity in the direction of the arrow in FIG. 19 (a) is not a sine wave as shown in FIG. 19 (b), but the hue and saturation obtained by HSV conversion of this RGB intensity are about π / 2 as shown in FIG. 19 (c). The phase difference is a sine wave.

  In Example 3 using the illumination pattern shown in FIG. 19, the number of times of imaging may be two times t = 0 and t = π. The hue and saturation of each of the two captured images are subjected to HSV conversion to be decomposed into a hue image and a saturation image, thereby obtaining a total of four images including two hue images and two saturation images. These four images are equivalent to the four images captured at t = 0, π / 2, π, and 3π / 2 in the second embodiment. That is, the four imaging operations in the second embodiment need only be performed twice in the third embodiment. FIG. 20 shows that two images can be obtained from this one image. From the illumination pattern at t = 0 in FIG. 20A, a hue image at t = 0 in FIG. 20A1 is obtained. A saturation image of t = π / 2 in FIG. 20 (a2) is obtained (both are expressed in luminance). Further, a hue image of t = π in FIG. 20B1 and a saturation image of t = 3π / 2 in FIG. 20B2 are obtained from the illumination pattern of t = π in FIG. 20B. Based on the difference between the amplitude and phase of the four images, an evaluation value is obtained in the same manner as in the second embodiment, and defective pixels are extracted, and a defect can be detected therefrom.

Example 4
The fourth embodiment is an example in which an illumination pattern in which saturation is modulated in a sine wave shape in the Y direction orthogonal to the X direction is used for a stripe pattern in which the hue is modulated in a sine wave shape in the X direction. This illumination pattern is as shown in FIG. The RGB intensity in the X direction indicated by the arrow in FIG. 21A does not have a sine wave shape as shown in FIG. 21B, but the hue and saturation obtained by HSV conversion of the RGB intensity in FIG. As shown in 21 (c), the hue changes like a sine wave, and the saturation is constant. In addition, as shown in FIG. 21D, the RGB intensity in the Y direction is constant in the G intensity, whereas the R and B intensity changes in a sine wave form. As shown in FIG. 21E, the hue and saturation obtained by HSV conversion of the RGB intensity are constant and the saturation changes in a sine wave shape.

  In the same way as in the second embodiment, the phase is displaced to t = 0, π / 2, π, and 3π / 2 by moving the illumination pattern, and imaging is performed four times. Since images of hue and saturation can be obtained from each captured image, eight images can be obtained from the four illumination patterns. FIG. 22 is a diagram showing that eight images can be obtained from four illumination patterns. FIG. 22 (a) is a diagram in which t = 0, π / 2, π, and 3π / 2 are displaced. 22 (b) shows four hue pattern images in which only the hue is extracted from FIG. 22 (a) and the luminance is expressed with constant saturation. FIG. 22C shows images of four saturation patterns in which the hue is constant and only the saturation is extracted from FIG. In this manner, eight images can be obtained from the four illumination patterns. Based on these eight images, defective pixels are extracted as compared with the reference image. Although the number of times of imaging is the same as that of the second embodiment, it is possible to detect a defect with higher accuracy.

  The embodiments of the surface defect inspection apparatus and the surface defect inspection method of the present invention have been described above. However, these embodiments are not limited to the above-described contents, and can be carried out in various modes without departing from the gist of the present invention. To get.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1)
An illumination unit that irradiates a reference surface and an inspection surface with an illumination pattern whose hue periodically changes in at least two directions on a plane;
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern;
A defect detection unit that detects the presence or absence of a defect on the inspection surface based on a difference in hue between both images captured by the imaging unit;
A surface defect inspection apparatus comprising:
(Appendix 2)
The illumination pattern is characterized in that a hue is modulated in a sine wave shape in a first direction of a plane, and a phase of the sine wave is periodically changed in a second direction perpendicular to the first direction. The surface defect inspection apparatus according to appendix 1.
(Appendix 3)
The illumination unit irradiates the illumination pattern by moving the illumination pattern to the reference surface and the inspection surface a plurality of times,
The imaging unit images the reference surface and the inspection surface for each movement,
The defect detection unit obtains a first evaluation value for a hue change from a plurality of images of the reference plane imaged by the imaging unit and obtains a second evaluation value for a hue change from the plurality of images of the inspection surface, The surface defect inspection apparatus according to appendix 2, wherein the presence or absence of a defect on the inspection surface is detected based on a difference between the first evaluation value and the second evaluation value.
(Appendix 4)
An illumination pattern in which saturation is modulated in a sine wave shape in the first direction of the plane and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction is inspected with the reference plane Illuminating unit that illuminates by moving multiple times to the surface,
An imaging unit that images the reference surface and the inspection surface for each movement;
A first evaluation value for a saturation change is obtained from a plurality of images of the reference plane imaged by the imaging unit, and a second evaluation value for a saturation change is obtained from a plurality of images of the inspection surface, and the first evaluation value is obtained. A defect detection unit that detects the presence or absence of a defect on the inspection surface based on the difference between the evaluation value and the second evaluation value;
A surface defect inspection apparatus comprising:
(Appendix 5)
Hue and saturation have a predetermined phase difference in the first direction of the plane and are modulated in a sine wave shape, and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction. An illumination unit that illuminates the reference surface and the inspection surface by moving the illumination pattern changed to
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern for each movement;
A first evaluation value for a hue change is obtained from a plurality of images of the reference surface imaged by the imaging unit, and a second evaluation value for a hue change is obtained from a plurality of images of the inspection surface, and the first evaluation is performed. A defect detection unit for detecting the presence or absence of a defect on the inspection surface based on a difference between the value and the second evaluation value;
A surface defect inspection apparatus comprising:
(Appendix 6)
Hue is modulated sinusoidally in the first direction of the plane, the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction, and the saturation is modulated in the sinusoidal shape An illumination unit that moves the illuminated pattern and illuminates the reference surface and the inspection surface;
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern for each movement;
A first evaluation value for a hue change is obtained from a plurality of images of the reference surface imaged by the imaging unit, and a second evaluation value for a hue change is obtained from a plurality of images of the inspection surface, and the first evaluation is performed. A defect detection unit for detecting the presence or absence of a defect on the inspection surface based on a difference between the value and the second evaluation value;
A surface defect inspection apparatus comprising:
(Appendix 7)
An illumination procedure for irradiating a reference surface and an inspection surface with an illumination pattern whose hue periodically changes in at least two directions on a plane;
An imaging procedure for imaging the reference surface and the inspection surface irradiated with the illumination pattern;
A defect detection procedure for detecting the presence or absence of defects on the inspection surface based on a difference in hue between both images captured by the imaging unit;
A surface defect inspection method characterized by comprising:
(Appendix 8)
The surface defect inspection apparatus according to appendix 3, wherein the movement amount of the plurality of movements is 1 / n (n ≧ 3) of the period of the sine wave, and the number of movements is n times.
(Appendix 9)
The illumination unit has a plurality of the illumination patterns having different phases in the first direction, and irradiates the illumination pattern while switching the illumination pattern to the reference surface and the inspection surface, respectively.
The imaging unit images the reference surface and the inspection surface every time the switching is performed,
The defect detection unit obtains a first evaluation value for a hue change from a plurality of images of the reference plane imaged by the imaging unit and obtains a second evaluation value for a hue change from the plurality of images of the inspection surface, The surface defect inspection apparatus according to appendix 2, wherein the presence or absence of a defect on the inspection surface is detected based on a difference between the first evaluation value and the second evaluation value.
(Appendix 10)
The phase is 1 / n (n ≧ 3) of the period of the sine wave,
The surface defect inspection apparatus according to appendix 9, wherein the illumination pattern is switched n times.

DESCRIPTION OF SYMBOLS 10 Inspection surface 11 Base layer 12 Clear layer 20 Foreign material 21 Defect 22 Scratch defect 23 Scratch defect 30 Illumination part 31 Bright part 32 Dark part 40 Imaging part 41 (Specular reflection in a part without a defect, and it injects into an imaging part) 42 ( Light reflected from the inclined surface of the defect and incident on the imaging unit) 50 Captured image 51 Bright part 52 Dark part 100 Surface defect inspection apparatus 101 Surface defect inspection apparatus 200 Control unit 201 Control unit 210 CPU
220 RAM
230 Image Data Storage Unit 231 Inspection Program Storage Unit 232 Inspection Program Storage Unit 240 Imaging Control Unit (Imaging CNT)
250 Lighting Control Unit (Lighting CNT)
260 Movement control unit (movement CNT)
270 IO / IF (Input Output / Interface)
271 Keyboard (KB)
272 Display (DISP)
DESCRIPTION OF SYMBOLS 300 Illumination part 301 Illumination part 310 Illumination pattern panel 320 Moving mechanism 400 Imaging part

Claims (6)

An illumination pattern in which the hue is modulated in a sine wave shape in the first direction of the plane and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction is defined as a reference surface and an inspection surface. And an illumination unit for irradiating
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern;
A defect detection unit that detects the presence or absence of a defect on the inspection surface based on a difference in hue between both images captured by the imaging unit;
A surface defect inspection apparatus comprising: The illumination unit irradiates the illumination pattern by moving the illumination pattern to the reference surface and the inspection surface a plurality of times,
The imaging unit images the reference surface and the inspection surface for each movement,
The defect detection unit obtains a first evaluation value for a hue change from a plurality of images of the reference plane imaged by the imaging unit and obtains a second evaluation value for a hue change from the plurality of images of the inspection surface, Based on the difference between the first evaluation value and the second evaluation value, the presence or absence of a defect on the inspection surface is detected.
The surface defect inspection apparatus according to claim 1. An illumination pattern in which saturation is modulated in a sine wave shape in the first direction of the plane and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction is inspected with the reference plane Illuminating unit that illuminates by moving multiple times to the surface,
An imaging unit that images the reference surface and the inspection surface for each movement;
A first evaluation value for a saturation change is obtained from a plurality of images of the reference plane imaged by the imaging unit, and a second evaluation value for a saturation change is obtained from a plurality of images of the inspection surface, and the first evaluation value is obtained. A defect detection unit that detects the presence or absence of a defect on the inspection surface based on the difference between the evaluation value and the second evaluation value;
A surface defect inspection apparatus comprising: Hue and saturation have a predetermined phase difference in the first direction of the plane and are modulated in a sine wave shape, and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction. An illumination unit that illuminates the reference surface and the inspection surface by moving the illumination pattern changed to
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern for each movement;
A first evaluation value for a hue change is obtained from a plurality of images of the reference surface imaged by the imaging unit, and a second evaluation value for a hue change is obtained from a plurality of images of the inspection surface, and the first evaluation is performed. A defect detection unit for detecting the presence or absence of a defect on the inspection surface based on a difference between the value and the second evaluation value;
A surface defect inspection apparatus comprising: Hue is modulated sinusoidally in the first direction of the plane, the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction, and the saturation is modulated in the sinusoidal shape An illumination unit that moves the illuminated pattern and illuminates the reference surface and the inspection surface;
An imaging unit that images the reference surface and the inspection surface irradiated with the illumination pattern for each movement;
A first evaluation value for a hue change is obtained from a plurality of images of the reference surface imaged by the imaging unit, and a second evaluation value for a hue change is obtained from a plurality of images of the inspection surface, and the first evaluation is performed. A defect detection unit for detecting the presence or absence of a defect on the inspection surface based on a difference between the value and the second evaluation value;
A surface defect inspection apparatus comprising: An illumination pattern in which the hue is modulated in a sine wave shape in the first direction of the plane and the phase of the sine wave is periodically changed in a second direction perpendicular to the first direction is defined as a reference surface and an inspection surface. Lighting procedure to irradiate and
An imaging procedure for imaging the reference surface and the inspection surface irradiated with the illumination pattern;
A defect detection procedure for detecting the presence or absence of a defect on the inspection surface based on the difference in hue between the two images captured in the imaging procedure;
A surface defect inspection method characterized by comprising: