Inspection system and inspection method

Abstract

The inspection system includes: a light-emitting device for illuminating an object; a collimating lens disposed between the light-emitting device and the object; and an imaging device for imaging the object. The light-emitting device varies the light-emitting position. The inspection system further includes an image analysis unit that generates an analysis image in which the value of each pixel corresponds to the normal direction of the surface of the object imaged in the corresponding pixel by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other. Thereby, the labor for adjustment for inspection can be reduced, and the detection accuracy of defects is improved.

Description

Technical Field

[0001] The present disclosure relates to an inspection system and an inspection method. Background Art

[0002] In the field of FA (Factory Automation) and the like, the following technique is known: while illuminating an object having a shiny surface such as metal and taking a picture, the obtained image is used to inspect the appearance of the object.

[0003] As a lighting method, parallel light coaxial illumination is known. By using parallel light coaxial illumination, when there are defects such as fine scratches and gentle unevenness on the surface of the object, a lightness shading distribution corresponding to the defects appears in the image. Therefore, by confirming the shading distribution, the presence or absence of defects can be inspected.

[0004] The reflection type phase shift method is disclosed in Utility Model Registration No. 3197766 (Patent Document 1). In the reflection type phase shift method, while moving the slit light by one cycle, the object is irradiated. Stripes are observed in the image obtained by taking a picture, and the brightness change is different depending on the presence or absence of defects. Therefore, by confirming the brightness change, defects can be detected. For example, for each position on the surface of the inspected object to be photographed, the maximum value and the minimum value of the image brightness in one cycle are obtained, and based on the difference image between the maximum value image collecting the maximum values at each position on the surface of the inspected object and the minimum value image collecting the minimum values, defects on the surface of the inspected object are detected.

[0005] An inspection lighting device having a filter unit that forms a plurality of solid angle regions having different light attributes as the irradiation solid angle of the light irradiated to each point of the object is disclosed in Japanese Patent No. 5866586 (Patent Document 2). By using this inspection lighting device, it is possible to set the shape, size, inclination of the irradiation solid angle of the light, and the solid angle region having a specific light attribute within the irradiation solid angle to be substantially uniform throughout the field of view. As a result, even minute defects and the like can be detected under substantially the same detection conditions.

[0006] Prior Art Documents

[0007] Patent Documents

[0008] Patent Document 1: Utility Model Registration No. 3197766

[0009] Patent Document 2: Japanese Patent No. 5866586 Summary of the Invention

[0010] Problems to be Solved by the Invention

[0011] In the case of using collimated coaxial illumination, the object needs to be set such that the optical axis of the collimated light is parallel to the normal direction of the surface of the object. Therefore, it takes time to adjust the setting of the object.

[0012] In the case of using the reflection type phase shift method described in Patent Document 1, if the diffuse reflectance of the surface of the object is large, the contrast of the observed fringes decreases, and thus the detection accuracy of defects decreases.

[0013] In the case of using the inspection illumination device described in Patent Document 2, it takes time to adjust the filter unit according to the reflection characteristics of the surface of the object.

[0014] The present disclosure has been made in view of the above problems, and an object thereof is to provide an inspection system and an inspection method capable of reducing the adjustment effort for inspection and having high detection accuracy of defects.

[0015] Means for Solving the Problem

[0016] According to an example of the present disclosure, an inspection system for inspecting the surface of an object includes: a light emitting device for illuminating the object; a collimating lens disposed between the light emitting device and the object; and a photographing device for photographing the object. The light emitting device makes the light emitting position variable. The inspection system further includes an image analysis unit that generates a first analysis image in which the value of each pixel corresponds to the normal direction of the surface of the object reflected in the corresponding pixel by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light emitting positions are different from each other.

[0017] According to the above disclosure, in the first analysis image, the values of the pixels that reflect defects such as unevenness or scratches where the normal direction of the surface changes are different from the values of other pixels. Therefore, by confirming the first analysis image, defects can be detected with high accuracy.

[0018] Furthermore, the relative positional relationship between the object and the photographing device only needs to be set such that the brightness shows a peak at any of the plurality of light emitting positions. Therefore, unlike Patent Document 1, it does not take time to adjust the setting of the object. Moreover, since it does not have a filter unit as described in Patent Document 2, it does not take time to adjust the filter unit.

[0019] As described above, according to the inspection system having the above structure, the adjustment effort for inspection can be reduced, and the detection accuracy of defects can be improved.

[0020] In the above disclosure, the value of each pixel of the first analysis image represents the phase of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emitting position.

[0021] The phase of the waveform representing the relationship between the luminance of a pixel and the light-emitting position depends on the normal direction of the surface of the object imaged in the corresponding pixel. Therefore, according to the above disclosure, the first analysis image can accurately represent the distribution of the normal direction of the surface of the object.

[0022] In the above disclosure, the value of each pixel of the first analysis image represents the light-emitting position where the luminance becomes a peak in the following waveform, which represents the relationship between the luminance of the corresponding pixel in a plurality of captured images and the light-emitting position.

[0023] The light-emitting position where the luminance becomes a peak depends on the normal direction of the surface of the object. Therefore, according to the above disclosure, the first analysis image can accurately represent the distribution of the normal direction of the surface of the object.

[0024] In the above disclosure, the image analysis unit further generates a second analysis image by analyzing a plurality of captured images. The value of each pixel of the second analysis image is the amplitude of the following waveform, which represents the relationship between the luminance of the corresponding pixel in a plurality of captured images and the light-emitting position.

[0025] According to the above disclosure, by confirming the second analysis image, it is possible to accurately detect dirt on a substance that reduces the degree of specular reflection of light, or fine scratches on the surface of a glass object.

[0026] In the above disclosure, the light-emitting device includes a plurality of light sources, and sequentially switches the light sources to be lit among the plurality of light sources. The plurality of light sources are arranged on a plane perpendicular to the optical axis of the collimating lens, or on a spherical surface having a center on the optical axis. According to the above disclosure, the light-emitting device can easily change the light-emitting position.

[0027] In the above disclosure, the image analysis unit further generates a composite image obtained by synthesizing a plurality of captured images.

[0028] According to the above disclosure, by confirming the composite image, it is possible to accurately detect uneven defects with low contrast of diffuse reflection without being affected by the surface unevenness of the object.

[0029] In the above disclosure, the distance between the light-emitting position and the collimating lens is equal to or greater than the focal length of the collimating lens.

[0030] According to the above disclosure, when the distance between the light-emitting position and the collimating lens is equal to the focal length of the collimating lens, it is possible to make the irradiation conditions on the surface of the object uniform. When the distance between the light-emitting position and the collimating lens is greater than the focal length of the collimating lens, the degree of freedom of the installation location of the imaging device is increased.

[0031] In the above disclosure, the inspection system further has a half mirror disposed between the collimating lens and the object. The light irradiated from the light emitting device and transmitted through the collimating lens is reflected by the half mirror and irradiated onto the object. The reflected light from the object is incident on the imaging device through the half mirror.

[0032] According to the above disclosure, the imaging device can easily image the object irradiated by the light from the light emitting device.

[0033] According to an example of the present disclosure, an inspection system for inspecting the surface of an object includes: a light emitting device for illuminating the object; a collimating lens disposed between the light emitting device and the object; and an imaging device for imaging the object. The light emitting device makes the light emitting position variable. The inspection system further has an image analysis unit that generates an analysis image by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light emitting positions are different from each other. The value of each pixel of the analysis image is a feature amount of a waveform as follows, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emitting position.

[0034] According to the above disclosure, the value of each pixel of the analysis image is a feature amount of a waveform as follows, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emitting position. The waveform representing the relationship between the pixel brightness and the light emitting position changes according to the normal direction of the surface of the object imaged in the pixel and the degree of regular reflection of light on the surface. Therefore, by confirming the analysis image with the feature amount of the waveform as the pixel value, it is possible to grasp the distribution of the normal direction of the surface of the object or the distribution of the degree of regular reflection of light on the surface of the object. As a result, it is possible to detect defects such as unevenness and scratches that affect the normal direction of the surface or dirt and scratches that affect the degree of regular reflection of light with high precision.

[0035] Furthermore, the relative positional relationship between the object and the imaging device only needs to be set so that the brightness shows a peak at any of the plurality of light emitting positions. Therefore, unlike in Patent Document 1, there is no need to spend time on setting and adjusting the object. Moreover, since it does not have a filter unit as described in Patent Document 2, there is no need to spend time on adjusting the filter unit.

[0036] As described above, according to the inspection system having the above structure, it is possible to reduce the time and effort for inspection adjustment and improve the defect detection accuracy.

[0037] According to an example of the present disclosure, an inspection method for the surface of an object to be inspected includes the following steps: irradiating the object with the light irradiated from a light-emitting device and transmitted through a collimating lens while photographing the object. The step of performing the photographing includes the following steps: switching the light-emitting position in the light-emitting device; and obtaining a plurality of captured images from multiple captures in a state where the light-emitting positions are different from each other. The inspection method further includes the following step: generating an analysis image in which the value of each pixel corresponds to the normal direction of the surface of the object imaged in the corresponding pixel by analyzing the plurality of captured images.

[0038] According to this disclosure, it is also possible to reduce the labor for adjustment in inspection and improve the detection accuracy of defects.

[0039] Effects of the Invention

[0040] According to the present disclosure, it is possible to reduce the labor for adjustment in inspection and improve the detection accuracy of defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram showing the overall structure of the inspection system 1 of the present embodiment.

[0042] Figure 2 is a schematic diagram showing the structure of the inspection system 1A of Specific Example 1.

[0043] Figure 3 is a schematic diagram showing the structure of the inspection system of Specific Example 2.

[0044] Figure 4 is a diagram showing an example of the hardware structure of the image analysis unit.

[0045] Figure 5 is a diagram showing an example of a plurality of captured images respectively obtained from multiple captures in a state where the light-emitting positions are different from each other.

[0046] Figure 6 is a diagram showing the captured image when light is emitted at the light-emitting position of X = 0 mm and Y = -3 mm.

[0047] Figure 7 is a diagram showing the brightness change of pixels 50 to 52 among the selected 7 captured images.

[0048] Figure 8 is shown according to Figure 5 a diagram of the normal image 20Y generated from the 7 captured images surrounded by solid lines.

[0049] Figure 9 is a diagram showing the differential image 28 generated by using the reflection type phase shift method described in Patent Document 1.

[0050] Figure 10is a diagram showing the direct reflection image 22Y generated from seven captured images surrounded by solid lines as shown in Figure 5

[0051] Figure 11 is a diagram showing the composite image 24 obtained by combining multiple captured images.

[0052] Figure 12 is a diagram showing 25 captured images respectively obtained from 25 captures in states where the light emission positions are different from each other, the normal images 20X, 20Y, the direct reflection image 22Y, and the composite image 24 generated from these captured images.

[0053] Figure 13 is a diagram showing an example of the generation processing flow of the analysis image.

[0054] Figure 14 is a diagram showing an example of a method for changing the light emission position in the light emitting device 10A.

[0055] Figure 15 is a diagram showing another example of a method for changing the light emission position in the light emitting device 10A.

[0056] Figure 16 is a diagram showing the solid angle of three-dimensional irradiation for each point in the inspection target area.

[0057] Figure 17 is a diagram showing the specular reflection light beam 90 of the end point P of the inspection target area 3.

[0058] Figure 18 is a diagram showing the specular reflection light beam 92 of the end point Q of the inspection target area 3.

[0059] Figure 19 is a diagram showing the arrangement area of the imaging device 16.

[0060] Figure 20 is a diagram showing the structure of the inspection system 1C of Modification 1.

[0061] Figure 21 is a diagram showing examples of four lighting states of the light emitting device 10C.

[0062] Figure 22 is a diagram showing the structure of the inspection system 1D of Modification 2.

[0063] Figure 23 is a diagram showing the optical path when the distance L between the light emitting device 10A and the collimator lens 12 is set to be longer than the focal length f of the collimator lens 12. DETAILED DESCRIPTION

[0064] ​The embodiments of the present invention will be described in detail with reference to the accompanying drawings. In addition, the same or corresponding parts in the drawings are denoted by the same reference numerals, and their descriptions will not be repeated.

[0065] §1 Application Example

[0066] Refer to Figure 1 , and the application example of the present invention will be described. Figure 1 FIG. is a schematic diagram showing the overall structure of the inspection system 1 of the present embodiment. The inspection system 1 inspects an inspection object area 3 on the surface of an object 2 to be inspected. The object 2 includes an object having a shiny surface such as metal or glass. The inspection system 1 is assembled on a production line, for example, to inspect the presence or absence of defects in the inspection object area 3. Defects include scratches, unevenness, dirt, and attachment of dust.

[0067] As Figure 1 shown, the inspection system 1 has a light emitting device 10, a collimating lens 12, a photographing device 16, and an image analysis unit 18 as main components. And, Figure 1 the illustrated inspection system 1 has a half mirror 14.

[0068] The light emitting device 10 is a device for illuminating the object 2. The light emitting device 10 makes the light emitting position variable. In Figure 1 the shown example, the light emitting device 10 can emit light from a plurality of light emitting positions (including light emitting positions PA, PB, PC) on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. The light emitting position PB is located on the optical axis 80 of the collimating lens 12.

[0069] The collimating lens 12 is disposed on the optical path between the light emitting device 10 and the object 2. In Figure 1 the shown example, the collimating lens 12 is disposed at a position separated from the imaginary plane 82 by a focal length f. Therefore, the light irradiated from any one of the light emitting positions (for example, light emitting positions PA, PB, PC) arranged on the imaginary plane 82 becomes parallel light through the collimating lens 12.

[0070] The half mirror 14 is disposed on the optical path between the collimating lens 12 and the object 2. As Figure 1 shown, the inspection object area 3 of the object 2 is substantially parallel to the optical axis 80 of the collimating lens 12. Therefore, the half mirror 14 is arranged such that the angle formed by the optical axis 80 of the collimating lens 12 and the reflecting surface is 45 degrees in order to direct the light emitted from the light emitting device 10 toward the object 2. Thus, the light irradiated from the light emitting device 10 and transmitted through the collimating lens 12 is reflected by the half mirror 14 and irradiates the object 2.

[0071] As an example, the imaging device 16 includes, in addition to an optical system such as a lens, imaging elements divided into a plurality of pixels such as a CCD (Coupled Charged Device) and a CMOS (Complementary Metal Oxide Semiconductor) sensor. The imaging device 16 is arranged on the side of the half mirror 14 opposite to the object 2 in such a manner that the inspection object area 3 of the object 2 is included in the field of view. Specifically, the imaging device 16 is arranged such that the optical axis 84 of the imaging device 16 is perpendicular to the optical axis 80 of the collimating lens 12, and the angle formed by the optical axis 84 of the imaging device 16 and the half mirror 14 is 45 degrees. Thus, the reflected light from the object 2 passes through the half mirror 14 and enters the imaging device 16. The imaging device 16 outputs image data obtained by imaging (hereinafter referred to as "captured image") to the image analysis unit 18.

[0072] The image analysis unit 18 analyzes a plurality of captured images respectively obtained from multiple captures in states where the light emission positions are different from each other.

[0073] The light emitted from one light emission position in the light emitting device 10 is reflected by the half mirror 14 after passing through the collimating lens 12 and irradiates the inspection object area 3 of the object 2. Since the collimating lens 12 is arranged between the light emission position and the object 2, the irradiation conditions (light quantity, irradiation angle, irradiation solid angle, etc.) at each point in the inspection object area 3 of the object 2 become uniform.

[0074] If the light emission position changes, the irradiation angle of the light irradiating each point in the inspection object area 3 changes. In Figure 1 the example shown, the light LA irradiated from the light emission position PA is reflected by the half mirror 14 after passing through the collimating lens 12 and irradiates the object 2 from the left side of the paper surface. The light LB irradiated from the light emission position PB is reflected by the half mirror 14 after passing through the collimating lens 12 and irradiates the object 2 along the vertical direction. The light LC irradiated from the light emission position PC is reflected by the half mirror 14 after passing through the collimating lens 12 and irradiates the object 2 from the right side of the paper surface.

[0075] When the object 2 is an object with a shiny surface such as metal or glass, most of the light irradiated onto the inspection target area 3 is specularly reflected. The amount of light (specular reflection component) that is specularly reflected and incident on the imaging device 16 at each point in the inspection target area 3 depends on the irradiation conditions at that point and the normal direction at that point. Therefore, if the irradiation angle of the light to each point in the inspection target area 3 changes, the specular reflection component incident on the imaging device 16 also changes. However, as described above, the irradiation conditions at each point in the inspection target area 3 are uniform. Therefore, at each point in the inspection target area 3, the light-emitting position where the specular reflection component incident on the imaging device 16 is the largest depends on the normal direction at that point. Therefore, in order to confirm the distribution of the normal directions at each point in the inspection target area 3, the image analysis unit 18 analyzes the brightness change of each pixel for a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other.

[0076] For example, when the inspection target area 3 is completely flat, that is, when the normal direction at each point in the inspection target area 3 is fixed, for the plurality of captured images, the brightness change of each pixel becomes uniform. On the other hand, when there are defects such as unevenness or scratches in a part of the inspection target area 3, the normal direction of the defect is different from the normal direction of the part other than the defect. Therefore, for the plurality of captured images, the brightness change of the pixels showing the defect is different from the brightness change of the pixels showing the part other than the defect.

[0077] In this way, by analyzing the brightness change of each pixel, the normal direction of the part imaged in the corresponding pixel can be determined. Therefore, the image analysis unit 18 generates an analysis image (hereinafter referred to as "normal image 20") in which the value of each pixel corresponds to the normal direction of the surface of the object 2 imaged in the corresponding pixel by analyzing a plurality of captured images. In Figure 1 In the shown normal image 20, the value of the pixel showing the defect F with a gentle depression formed is different from the value of other pixels (that is, the pixels showing the flat part without defects). Thus, by confirming the normal image 20, defects such as unevenness and scratches where the normal direction changes can be detected with high accuracy.

[0078] The relative positional relationship between the object 2 and the imaging device 16 only needs to be set so that the brightness shows a peak at any of the plurality of light-emitting positions. Therefore, it is not necessary to spend time on setting and adjusting the object 2 as in the case of using coaxial illumination with parallel light.

[0079] Furthermore, the inspection system 1 of the present embodiment does not have a filter unit as described in Patent Document 2 that forms a plurality of solid angle regions with different light attributes as the irradiation solid angle of the light irradiated onto each point of the object. Therefore, it is not necessary to spend time on adjusting the filter unit.

[0080] As described above, the inspection system 1 according to the present embodiment can reduce the labor for adjustment in inspection and can improve the detection accuracy of defects.

[0081] §2 Specific Examples

[0082] <A. Specific Example 1 of the Inspection System>

[0083] Figure 2 It is a schematic diagram showing the structure of the inspection system 1A of Specific Example 1. As Figure 2 shown, the inspection system 1A includes a light emitting device 10A, a collimating lens 12, a half mirror 14, a photographing device 16, and an image analysis unit 18.

[0084] The light emitting device 10A includes a plurality of light sources 101. The light source 101 can be a point light source or a surface light source.

[0085] The plurality of light sources 101 are arranged on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. The distance L between the imaginary plane 82 and the collimating lens 12 is the same as the focal length f of the collimating lens 12.

[0086] In Figure 2 the example shown, 49 light sources 101 are arranged in a matrix of 7 rows and 7 columns in the X direction. The X direction is a direction perpendicular to the optical axis 80 of the collimating lens 12 (vertical direction in Figure 2 ), and the Y direction is a direction perpendicular to the X direction and perpendicular to the optical axis 80 of the collimating lens 12.

[0087] The light emitting device 10A changes the light emitting position by sequentially switching the light sources 101 to be lit among the plurality of light sources 101.

[0088] <B. Specific Example 2 of the Inspection System>

[0089] Figure 3 It is a schematic diagram showing the structure of the inspection system of Specific Example 2. As Figure 3 shown, the inspection system 1B includes a light emitting device 10B, a collimating lens 12, a half mirror 14, a photographing device 16, and an image analysis unit 18.

[0090] The light emitting device 10B includes: one light source 102; and an XY stage 103 that moves the light source 102 in the X direction and the Y direction on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. The light source 102 can be a point light source or a surface light source.

[0091] The light emitting device 10B changes the light emitting position by moving the XY stage 103.

[0092] <C. Hardware Structure of the Image Analysis Unit>

[0093] The image analysis unit 18 typically has a structure that follows a general computer architecture, and realizes various processes by a processor executing a pre-installed program.

[0094] Figure 4 It is a diagram showing an example of the hardware structure of the image analysis unit. As Figure 4 shown, the image analysis unit 18 includes a processor 180 such as a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit), a RAM (Random Access Memory) 181, a display controller 182, a system controller 183, an I / O (Input Output) controller 184, a hard disk 185, a camera interface 186, an input interface 187, a communication interface 189, and a memory card interface 190. These respective parts are centered around the system controller 183 and are connected in a manner capable of mutually performing data communication.

[0095] The processor 180 realizes target arithmetic processing by exchanging programs (codes) etc. with the system controller 183 and executing these programs (codes) etc. in a prescribed order.

[0096] The system controller 183 is respectively connected to the processor 180, the RAM 181, the display controller 182, and the I / O controller 184 via a bus, exchanges data etc. with each part, and is responsible for the overall processing of the image analysis unit 18.

[0097] Typically, the RAM 181 is a volatile storage device such as a DRAM (Dynamic Random Access Memory), and holds programs read from the hard disk 185, captured images received from the imaging device 16, processing results for the captured images, and workpiece data etc.

[0098] The display controller 182 is connected to the display device 5, and outputs a signal for displaying various information to the display device 5 in accordance with an internal command from the system controller 183. As an example, the display device 5 includes a liquid crystal display, an organic EL (Electro Luminescence) display, etc.

[0099] The I / O controller 184 controls data exchange with a recording medium or an external device connected to the image analysis unit 18. More specifically, the I / O controller 184 is connected to the hard disk 185, the camera interface 186, the input interface 187, the communication interface 189, and the memory card interface 190.

[0100] The hard disk 185 is typically a non-volatile magnetic storage device that stores the analysis program 191 executed by the processor 180 and the like. The analysis program 191 installed on the hard disk 185 is circulated in a state stored in the memory card 6 or the like. In addition, the captured images are stored in the hard disk 185. In addition, a semiconductor storage device such as a flash memory or an optical storage device such as a DVD-RAM (Digital Versatile Disk Random Access Memory) may be used instead of the hard disk 185.

[0101] The camera interface 186 serves as an input unit that receives the input of the captured image generated by the object 2 to be photographed, and mediates the data transmission between the processor 180 and the photographing device 16. More specifically, a photographing instruction is output from the processor 180 to the photographing device 16 via the camera interface 186. Thereby, the photographing device 16 photographs the subject and outputs the generated captured image to the processor 180 via the camera interface 186.

[0102] The input interface 187 mediates the data transmission between the processor 180 and the input device 7 such as a keyboard, a mouse, a touch panel, and a dedicated console. That is, the input interface 187 receives the operation instruction given by the user operating the input device 7.

[0103] The communication interface 189 mediates the data transmission between the processor 180 and other personal computers, server devices, etc. not shown. The communication interface 189 is typically composed of Ethernet (registered trademark), USB (Universal Serial Bus), etc. In addition, as described later, instead of installing the program stored in the memory card 6 into the image analysis unit 18, the program downloaded from a distribution server or the like may be installed into the image analysis unit 18 via the communication interface 189.

[0104] The memory card interface 190 mediates the data transmission between the processor 180 and the memory card 6 as a recording medium. That is, the analysis program 191 and the like executed by the image analysis unit 18 are circulated in a state stored in the memory card 6, and the memory card interface 190 reads out the analysis program 191 from the memory card 6. In addition, the memory card interface 190 writes the captured image obtained by the photographing device 16 and / or the processing result in the image analysis unit 18 into the memory card 6 in response to an internal instruction of the processor 180. In addition, the memory card 6 is composed of a general semiconductor storage device such as SD (Secure Digital), a magnetic recording medium such as a flexible disk, or an optical recording medium such as a CD-ROM (Compact Disk Read Only Memory).

[0105] In the case of using a computer having a structure following a general computer architecture as described above, in addition to the applications for providing the functions of the present embodiment, an OS (Operating System) for providing the basic functions of the computer can also be installed. In this case, the program of the present embodiment can also execute processing by calling necessary modules in the program modules provided as part of the OS in a prescribed order and / or timing. That is, the program of the present embodiment itself may sometimes not include the above-described modules, but cooperate with the OS to execute processing.

[0106] Furthermore, the analysis program 191 of the present embodiment can also be incorporated as part of other programs. In this case, the program itself does not include the modules included in the other programs having the above-described combination, but cooperates with the other program to execute processing. That is, the analysis program 191 as the present embodiment can also be in a form incorporated into other programs.

[0107] Alternatively, part or all of the functions provided by the execution of the analysis program 191 can be installed as a dedicated hardware circuit.

[0108] <D. Image Analysis Method>

[0109] Refer to Figures 5 to 13 to describe the image analysis processing of the image analysis unit 18. Figure 5 is a diagram showing an example of a plurality of captured images respectively obtained from multiple captures in states where the light emission positions are different from each other. Figure 5 shows the captured images respectively obtained when the light source 102 of the light emitting device 10B is moved to the light emission position at 7×7 = 49 at the 7th row in the X direction and the 7th column in the Y direction. In addition, the pitch of the light emission positions in the X direction is 1 mm, and the pitch of the light emission positions in the Y direction is 1 mm. The light emission position of the collimating lens 12 on the optical axis 80 is X = 0 mm and Y = 0 mm.

[0110] As Figure 5 shown, by making the light emission positions different, the brightness of each pixel of the captured image changes. The image analysis unit 18 generates an analysis image by analyzing the change in the brightness of each pixel. Therefore, the image analysis unit 18 can also select 7 captured images of 1 row or 1 column with a large change in brightness from the 49 captured images and analyze the selected 7 captured images. In the Figure 5 shown example, compared with the case where the light emission position changes in the X direction, the change in the brightness of each pixel is larger in the case where the light emission position changes in the Y direction. Therefore, the image analysis unit 18 selects 7 captured images in which the light emission position changes in the Y direction from the 49 captured images ( Figure 5The image surrounded by solid lines). The seven selected captured images are images captured when light emission occurs at light emission positions where X = 0 mm and Y = -3, -2, -1, 0, 1, 2, 3 mm. In addition, the selection of the analysis target image from among the multiple captured images can also be performed according to the user's input. Alternatively, it is also possible to automatically select the captured image of one row or one column with the largest change in brightness from among the multiple captured images.

[0111] Figure 6 is a diagram showing the captured image when light emission occurs at the light emission position where X = 0 mm and Y = -3 mm. As Figure 6 shown, the brightness of each pixel of the captured image becomes a value corresponding to the normal direction of the surface of the object 2 reflected in the corresponding pixel. In Figure 6 the example shown, the brightness increases in the order of pixel 51, pixel 52, and pixel 50. In addition, pixel 51 is a pixel that reflects the defect F which is a gentle depression. Pixel 50 is a pixel that reflects the surrounding part of the defect F. Pixel 52 is a pixel that reflects the flat part.

[0112] Figure 7 is a diagram showing the brightness change of pixels 50 to 52 among the seven selected captured images. In Figure 7 it, the horizontal axis represents the light emission position n when the captured image is captured. n = 0 to 6 respectively correspond to the light emission positions where Y = -3, -2, -1, 0, 1, 2, 3 mm.

[0113] As Figure 7 shown, the changes in brightness in pixels 50 and 52 that do not reflect the defect F are similar to each other. On the other hand, the change in brightness in pixel 51 that reflects the defect F is different from the changes in brightness in pixels 50 and 52 that do not reflect the defect F. Specifically, the waveform phase representing the relationship between the brightness in pixel 51 and the light emission position is different from the waveform phase representing the relationship between the brightness in pixels 50 and 52 and the light emission position. In other words, the light emission position where the brightness is maximum in pixel 51 is different from the light emission positions where the brightness is maximum in pixels 50 and 52. This is because the normal direction of the surface is different between the defect F which is a gentle depression and the part other than the defect F.

[0114] In this way, the phase of the waveform representing the relationship between the brightness of each pixel and the light emission position n depends on the normal direction of the surface of the object 2 reflected in the corresponding pixel. The image analysis unit 18 utilizes this point to analyze the seven selected captured images, thereby generating a normal image 20Y in which the value of each pixel corresponds to the normal direction of the surface of the object 2 reflected in the corresponding pixel.

[0115] The image analysis unit 18 performs a discrete Fourier transform on the waveform representing the relationship between the luminance of each pixel and the light emission position n, and obtains the phase of the component with frequency 1. Specifically, for each pixel, the image analysis unit 18 performs sine wave fitting using the following equation (1) on the waveform representing the relationship between the luminance of the corresponding pixel and the light emission position n, and calculates the phase. In Equation (1), N represents the number of captured images. For example, when Figure 5 the seven captured images enclosed by the solid line as shown are selected, N = 7. I n represents the luminance in the captured image when light is emitted at the light emission position n.

[0116] [Mathematical formula 1]

[0117]

[0118] The image analysis unit 18 generates a normal image 20Y with the phase as the pixel value. The phase corresponds to the normal direction of the surface of the object 2 imaged in the pixel.

[0119] Figure 8 is a diagram showing the normal image 20Y generated based on Figure 5 the seven captured images enclosed by the solid line as shown. Figure 9 is a diagram showing the differential image 28 generated using the reflection type phase shift method described in Patent Document 1. The differential image 28 represents the difference between the maximum value image that collects the maximum values of each pixel and the minimum value image that collects the minimum values of each pixel when the slit light is moved one cycle while irradiating the object 2. In Figure 8 and Figure 9 , in the region surrounded by the frame line 60, a defect F in the form of a gentle depression is imaged.

[0120] As Figure 9 shown, a difference can be seen between the pixel values of the pixels imaging the defect F and the pixel values of the parts other than the defect F. However, this difference is very small. This is because, due to the large diffuse reflectance of the surface of the object 2, the noise component caused by diffuse reflection increases in the differential image 28.

[0121] As Figure 8 shown, in the normal image 20Y, a relatively large difference can be seen between the pixel values of the pixels imaging the defect F and the pixel values of the parts other than the defect F. That is, in the normal image 20Y, the defect F is clearly shown. Therefore, by confirming the normal image 20Y, it is possible to detect with high precision the defect F in the form of a gentle depression that is difficult to detect in the inspection using the reflection type phase shift method described in Patent Document 1.

[0122] In addition, as the value of each pixel of the normal image 20Y, the image analysis unit 18 may also use, instead of the phase calculated by Equation (1), the value of the light emission position n at which the luminance becomes a peak in the following waveform, the waveform representing the relationship between the luminance of the corresponding pixel in a plurality of captured images and the light emission position n. In this case, the value of each pixel of the normal image 20Y also corresponds to the normal direction of the surface of the object 2 imaged in the corresponding pixel.

[0123] The image analysis unit 18 may also generate other analysis images on the basis of or instead of the normal image 20Y. For example, the image analysis unit 18 may calculate the amplitude of the waveform representing the relationship between the luminance of each pixel and the light emission position n, and generate an analysis image (hereinafter referred to as a "direct reflection image") having the amplitude as the value of each pixel.

[0124] The more the specular reflection component in the light irradiated onto the surface of the object 2 (in other words, the less the diffuse reflection component), the larger the amplitude of the waveform. In order for the specularly reflected light to enter the imaging device 16, the light emission position in the light emitting device 10, the normal direction of the surface, and the optical axis 84 of the imaging device 16 need to satisfy a specified condition. Therefore, in the pixels imaged on the surface where the specular reflection component in the irradiated light is large, the change in luminance when the light emission position is changed increases. As a result, the amplitude increases. On the contrary, in the pixels imaged on the surface where the diffuse reflection component in the irradiated light is large, the change in luminance when the light emission position is changed decreases. As a result, the amplitude decreases. Therefore, the direct reflection image having the amplitude as the value of each pixel represents the distribution of the degree of specular reflection of light on the surface of the object 2 imaged in the image.

[0125] Specifically, the image analysis unit 18 calculates the amplitude A of the waveform representing the relationship between the luminance of each pixel and the light emission position n according to the following Equation (2). The image analysis unit 18 generates a direct reflection image having the calculated amplitude A as the value of the pixel.

[0126] [Equation 2]

[0127]

[0128] Figure 10 represents the direct reflection image 22Y generated from the seven captured images surrounded by the solid line as shown according to Figure 5 . In the direct reflection image 22Y shown in Figure 10 , compared with the normal image 20Y shown in Figure 8 , no change in the pixel value caused by the defect F which is a gentle depression is observed. This is because the degree of specular reflection of light in the defect F is not different from the degree of specular reflection of light in the portion other than the defect F.

[0129] InFigure 10 In the region of the direct reflection image 22Y shown, which is surrounded by the frame line 62, text formed by a marker pen can be confirmed. In the part where the ink of the marker pen is attached, the degree of regular reflection of light is reduced. Therefore, the value of the pixel corresponding to this part is reduced in the direct reflection image 22Y. On the other hand, Figure 8 in the normal image 20Y shown, no part where the ink of the marker pen is attached is confirmed. This is because the ink is thinly attached to the surface of the object 2, and thus does not affect the normal direction of the surface of the object 2.

[0130] In this way, when dirt of a substance (such as ink) that reduces the degree of regular reflection of light adheres to the surface of the object 2, by confirming the direct reflection image 22Y, the dirt can be detected with high precision.

[0131] In Figures 5 to 10 the example, the image analysis unit 18 generates the normal image 20Y and the direct reflection image 22Y based on seven captured images in which the light-emitting position is changed in the Y direction. However, the image analysis unit 18 may also generate the normal image and the direct reflection image based on seven captured images in which the light-emitting position is changed in the X direction.

[0132] Moreover, the image analysis unit 18 may also generate a composite image obtained by synthesizing a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different. Specifically, for each pixel, the image analysis unit 18 calculates a value B represented by the following formula (3). The image analysis unit 18 generates a composite image in which the calculated value B is used as the pixel value.

[0133] [Formula 3]

[0134]

[0135] Figure 11 is a diagram showing the composite image 24 obtained by synthesizing a plurality of captured images. Figure 11 shows the composite image 24 obtained by synthesizing Figure 5 the 49 captured images shown. The composite image 24 is identical to the image captured when all the light sources 101 emit light simultaneously at all the light-emitting positions. By confirming the composite image 24, uneven defects with low contrast that undergo diffuse reflection can be detected with high precision without being affected by the surface unevenness of the object 2. In addition, when using Figure 2 the light-emitting device 10A shown, it is also possible to perform shooting by lighting all the light sources 101 simultaneously to obtain the composite image 24.

[0136] Thus, analysis images (normal line images, direct reflection images, composite images) generated by analyzing multiple captured images obtained from multiple captures in states where the light emission positions are different from each other are effective for defect detection. Depending on the type of defect to be detected, an analysis image to be used for inspection is appropriately selected from the normal line image, the direct reflection image, and the composite image.

[0137] The direct reflection image is effective not only for detecting dirt on substances that reduce the degree of regular reflection of light, but also for detecting fine scratches on the surface of, for example, a glass object 2.

[0138] Figure 12 FIG. shows 25 captured images respectively obtained from 25 captures in states where the light emission positions are different from each other, the normal line images 20X, 20Y, the direct reflection image 22Y, and the composite image 24 generated from these captured images. Figure 12 FIG. shows the captured images obtained when the light source 102 of the light emitting device 10B is moved to the 5×5 = 25 light emission positions that are 5 rows in the X direction and 5 columns in the Y direction. The normal line image 20X is generated from 5 captured images when light emission occurs at light emission positions where Y = 0 mm and X = -2, -1, 0, 1, 2 mm. The normal line image 20Y and the direct reflection image 22Y are generated from 5 captured images when light emission occurs at light emission positions where X = 0 mm and Y = -2, -1, 0, 1, 2 mm. The composite image 24 is generated from 25 captured images.

[0139] As Figure 12 shown, by magnifying a part of the direct reflection image 22Y, a fine scratch, i.e., a defect F1, formed on the surface of the glass object 2 is confirmed. On the other hand, even when the normal line images 20X, 20Y, and the composite image 24 are magnified, the defect F1 cannot be clearly confirmed. Thus, the direct reflection image 22Y is effective for detecting fine scratches on the surface of the glass object 2.

[0140] The image analysis unit 18 can also generate other analysis images by performing image processing on the normal line image and the direct reflection image.

[0141] Figure 13 FIG. is an example of a flowchart showing the generation process of the analysis image. As Figure 13 shown, the image analysis unit 18 acquires a plurality of captured images obtained by performing multiple captures while changing the light emission position in the X direction and the Y direction (step S1).

[0142] Next, the image analysis unit 18 generates a normal image 20X and a direct reflection image 22X by analyzing a plurality of captured images when the light emission position is changed in the X direction (step S2). Further, the image analysis unit 18 generates a normal image 20Y and a direct reflection image 22Y by analyzing a plurality of captured images when the light emission position is changed in the Y direction (step S3). The image analysis unit 18 generates a composite image 24 by synthesizing a plurality of captured images when the light emission position is changed in the X direction and the Y direction (step S4).

[0143] Next, the image analysis unit 18 generates a differential image 21X by applying a differential filter in the X direction to the normal image 20X (step S5). Further, the image analysis unit 18 generates a differential image 21Y by applying a differential filter in the Y direction to the normal image 20Y (step S6).

[0144] Next, the image analysis unit 18 generates an image 23X by multiplying the differential image 21X by the direct reflection image 22X (step S7). Further, the image analysis unit 18 generates an image 23Y by multiplying the differential image 21Y by the direct reflection image 22Y (step S8).

[0145] Next, the image analysis unit 18 generates a concavo-convex image 25 by adding the image 23X and the image 23Y (step S9). Then, the image analysis unit 18 performs binarization processing on the concavo-convex image 25 to generate a binary image 26 (step S10).

[0146] In Figure 13 the example shown, an analysis image for inspection is selected from the normal images 20X, 20Y, differential images 21X, 21Y, direct reflection images 22X, 22Y, images 23X, 23Y, composite image 24, concavo-convex image 25, and binary image 26 according to the type of defect to be detected.

[0147] For example, in the case of a defect having a concavo-convex shape, the normal directions of the defect and the portion other than the defect are different. Therefore, by using the normal images 20X, 20Y, the defect can be detected with high accuracy. Or, in the case of a defect having a concavo-convex shape, the normal direction changes sharply at the boundary between the defect and the portion other than the defect. Therefore, by using the differential images 21X, 21Y, the defect can be detected with high accuracy.

[0148] Depending on the surface state of the object 2, the brightness change in the captured image when the light-emitting position changes in the X direction may be quite different from the brightness change in the captured image when the light-emitting position changes in the Y direction. For example, in the case of the object 2 with fine lines formed on its surface, the brightness change in the captured image when the light-emitting position changes in the X direction is quite different from the brightness change in the captured image when the light-emitting position changes in the Y direction. In such a case, by adding the images 23X and 23Y to eliminate the influence caused by the fine lines, the images 23X and 23Y are respectively generated by multiplying the differential images 21X and 21Y with the direct reflection images 22X and 22Y. That is, in the values of the respective pixels of the uneven image 25, only the change in the normal direction caused by the defect is reflected. Thus, even for the object 2 with fine lines formed on its surface, by using the uneven image 25, it is possible to detect the defect with an uneven shape with high precision. Or, by using the binary image 26 generated by binarizing the uneven image 25, it is also possible to detect the defect with an uneven shape with high precision.

[0149] <E. Method for Changing the Light-Emitting Position of the Light-Emitting Device 10A>

[0150] In the case of using the light-emitting device 10A (refer to Figure 2 ) including a plurality of light sources 101, the light-emitting device 10A changes the light-emitting position by switching the light source 101 to be lit among the plurality of light sources 101. For example, the light-emitting device 10A only lights one of the plurality of light sources 101 and sequentially switches the light source 101 to be lit. Or, the light-emitting device 10A can also light at least two of the plurality of light sources 101 simultaneously and sequentially switch the light sources 101 to be lit simultaneously.

[0151] Figure 14 is a diagram showing an example of the method for changing the light-emitting position in the light-emitting device 10A. As Figure 14 shown, the light-emitting device 10A lights 7 light sources 101 arranged in one row along the X direction or one column along the Y direction simultaneously. In Figure 14 , (a) to (g) respectively show the captured images when 7 light sources 101 arranged at Y = -3, -2, -1, 0, 1, 2, 3 mm are lit simultaneously. In Figure 14 , (h) to (n) respectively show the captured images when 7 light sources 101 arranged at X = -3, -2, -1, 0, 1, 2, 3 mm are lit simultaneously.

[0152] Figure 15 is a diagram showing another example of the method for changing the light-emitting position in the light-emitting device 10A. As Figure 15 shown, the light-emitting device 10A lights 21 light sources 101 arranged in 3 rows along the X direction or 3 columns along the Y direction simultaneously. InFigure 15 Among them, (a), (b), (c), (d), and (e) respectively represent the captured images when 21 light sources 101 arranged at Y = -3 to -1 mm, -2 to 0 mm, -1 to 1 mm, 0 to 2 mm, and 1 to 3 mm are lit simultaneously. In Figure 15 Among them, (f), (g), (h), (i), and (j) respectively represent the captured images when 21 light sources 101 arranged at X = -3 to -1 mm, -2 to 0 mm, -1 to 1 mm, 0 to 2 mm, and 1 to 3 mm are lit simultaneously.

[0153] As Figure 14 and Figure 15 shown, according to the number of simultaneously lit light sources 101, the overall brightness of the captured image is different, and the number of captured images obtained is also different. Therefore, according to the surface state of the object 2 and the type of defect to be detected, the number of light sources 101 to be simultaneously lit can be selected.

[0154] <F. Arrangement Location of the Imaging Device 16>

[0155] The normal images 20X and 20Y are generated using the specularly reflected light at each point in the inspection target area 3 of the object 2. Therefore, when emitting light at at least one of the plurality of light emission positions, it is preferable that the imaging device 16 receives the specularly reflected light at each point in the inspection target area 3.

[0156] Figure 16 is a diagram showing the solid irradiation angle at each point in the inspection target area. Figure 16 shows the irradiation solid angle ω at the endpoints P and Q of the inspection target area 3 when all the light sources 101 of the light emitting device 10A are lit. The irradiation solid angle ω is represented by the following formula (4) using the length D of the light emitting device 10A in the direction perpendicular to the optical axis 80 of the collimating lens 12 and the focal length f.

[0157] ω = atan(D / f) Formula (4)

[0158] Since the collimating lens 12 is arranged between the light source 102 and the object 2, the irradiation solid angle ω is the same at each point in the inspection target area 3.

[0159] Figure 17 is a diagram showing the light beam 90 of the specularly reflected light at the endpoint P of the inspection target area 3. Figure 18 is a diagram showing the light beam 92 of the specularly reflected light at the endpoint Q of the inspection target area 3. Figure 17 and Figure 18 show the light beams of the specularly reflected light when all the light sources 101 of the light emitting device 10A are lit. As Figure 17 and Figure 18As shown, the light beams 90 and 92 of the specularly reflected light have the same solid angle as the illumination solid angle ω. By arranging the imaging device 16 in the passing area of the light beam 90, the specularly reflected light of the end point P can be received when any light source 101 is lit. By arranging the imaging device 16 in the passing area of the light beam 92, the specularly reflected light of the end point Q can be received when any light source 101 is lit.

[0160] Figure 19 It is a diagram showing the arrangement area of the imaging device 16. When the optical system of the imaging device 16 is a pinhole lens, the imaging device 16 is arranged in the area 94 where the light beam 90 of the specularly reflected light of the end point P and the light beam 92 of the specularly reflected light of the end point Q overlap. Thus, when light emission is performed at at least one of the plurality of light emission positions, the imaging device 16 can receive the specularly reflected light for all points on the inspection target area 3.

[0161] When the size of the optical system of the imaging device 16 is the same as that of the collimating lens 12, the imaging device 16 is arranged such that the position of the optical system coincides with the position of the virtual image 96 of the collimating lens 12, and the position of the imaging element of the imaging device 16 coincides with the position of the virtual image 802 of the light emitting device 10A. Thus, the imaging device 16 can receive the specularly reflected light from all points on the inspection target area 3.

[0162] When the size of the optical system of the imaging device 16 is smaller than that of the collimating lens 12 (for example, when the optical system of the imaging device 16 is a pinhole lens), the light emission position where the brightness becomes the peak may vary according to the position of the points on the inspection target area 3. Therefore, even if the inspection target area 3 is completely flat, in the normal images 20X and 20Y, the values of each pixel will change (error). However, the degree of this change is smaller than the degree of change caused by defects such as unevenness and scratches. Therefore, defects can be detected by confirming the differential images 21X and 21Y obtained by differentiating the normal images 20X and 20Y or the normal images 20X and 20Y respectively. In addition, as described later, by making the distances L between the light emitting devices 10A and 10B and the collimating lens 12 greater than the focal length f of the collimating lens 12, the above error can be reduced.

[0163] When the normal direction of the inspection target area 3 is distributed within a certain error range, in order to receive the specularly reflected light of each point on the inspection target area 3, the area where the imaging device 16 can be arranged is further restricted according to this error range.

[0164] When the imaging device 16 has a telecentric optical system, it is only necessary to arrange the imaging device 16 such that the telecentric optical system overlaps with the area 94. Therefore, the degree of freedom in arranging the imaging device 16 is increased.

[0165] <G. Function / Effect>

[0166] As described above, the inspection system 1 (1A, 1B) inspects the surface of the object 2. The inspection system 1 (1A, 1B) includes: a light-emitting device 10 (10A, 10B) for illuminating the object 2; a collimating lens 12 disposed between the light-emitting device 10 (10A, 10B) and the object 2; and an imaging device 16 for imaging the object 2. The light-emitting device 10 (10A, 10B) varies the light-emitting position. The inspection system 1 (1A, 1B) further includes an image analysis unit 18 that analyzes a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other. The image analysis unit 18 generates a normal image 20 (20X, 20Y) in which the value of each pixel corresponds to the normal direction of the surface of the object 2 reflected in the corresponding pixel.

[0167] In the normal image 20 (20X, 20Y), the values of the pixels that reflect defects such as unevenness or scratches where the normal direction of the surface changes are different from the values of other pixels. Thus, by confirming the normal image 20 (20X, 20Y), defects can be detected with high precision.

[0168] Furthermore, the relative positional relationship between the object 2 and the imaging device 16 only needs to be set such that the brightness shows a peak at any of the multiple light-emitting positions. Therefore, it does not take much effort to adjust the relative positional and postural relationship between the object 2 and the imaging device 16. And since the inspection system 1 (1A, 1B) does not have a filter unit as described in Patent Document 2, it does not take much effort to adjust the filter unit.

[0169] As described above, according to the inspection system 1 (1A, 1B), the effort for inspection adjustment can be reduced, and the detection accuracy of defects can be improved.

[0170] The value of each pixel of the normal image 20 (20X, 20Y) represents the phase of the following waveform The waveform represents the relationship between the brightness of the corresponding pixel in the multiple captured images and the light-emitting position. As Figure 7 shown, the phase depends on the normal direction of the surface of the object 2 reflected in the pixel. Therefore, the normal image 20 (20X, 20Y) can accurately represent the distribution of the normal direction of the surface of the object 2.

[0171] The value of each pixel of the normal image 20 (20X, 20Y) may also represent the light-emitting position at which the brightness becomes a peak in the following waveform, the waveform representing the relationship between the brightness of the corresponding pixel in the multiple captured images and the light-emitting position. As Figure 7As shown, the light-emitting position where the luminance becomes peak depends on the normal direction of the surface of the object 2. Therefore, the normal images 20 (20X, 20Y) can accurately represent the distribution of the normal direction of the surface of the object 2.

[0172] The image analysis unit 18 may also further generate direct reflection images 22X, 22Y by analyzing a plurality of captured images. The value of each pixel of the direct reflection images 22X, 22Y is the amplitude A of the waveform representing the relationship between the luminance and the light-emitting position of the corresponding pixel in the plurality of captured images.

[0173] By checking the direct reflection images 22X, 22Y, it is possible to accurately detect Figure 10 substances (such as ink) that reduce the degree of regular reflection of light, as shown. Furthermore, by checking the direct reflection images 22X, 22Y, it is possible to accurately detect Figure 12 fine scratches on the surface of the glass object 2, as shown.

[0174] The light-emitting device 10A includes a plurality of light sources 101, and sequentially switches the light source to be lit among the plurality of light sources 101. The plurality of light sources 101 are arranged on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. Thus, the light-emitting device 10A can easily change the light-emitting position.

[0175] The image analysis unit 18 may also further generate a composite image 24 obtained by synthesizing a plurality of captured images. By checking the composite image 24, it is possible to accurately detect low-contrast uneven defects that undergo diffuse reflection without being affected by the surface unevenness of the object 2.

[0176] In addition, the image analysis unit 18 may also generate at least one of the normal images 20 (20X, 20Y) and the direct reflection images 22X, 22Y. That is, the image analysis unit 18 generates an analysis image by analyzing a plurality of captured images respectively obtained from multiple shootings in states where the light-emitting positions are different from each other. The value of each pixel of the analysis image (including at least one of the normal images 20 (20X, 20Y) and the direct reflection images 22X, 22Y) is a feature quantity (including phase and amplitude) of the waveform representing the relationship between the luminance and the light-emitting position of the corresponding pixel in the plurality of captured images.

[0177] <H. Modification Example>

[0178] In Figure 2 Specific Example 1 shown, the plurality of light sources 101 included in the light-emitting device 10A are arranged in a matrix. However, the arrangement of the plurality of light sources 101 is not limited to a matrix.

[0179] Figure 20This is a diagram showing the structure of the inspection system 1C of Modification Example 1. As Figure 20 shown, the inspection system 1C includes a light-emitting device 10C, a collimating lens 12, a half mirror 14, a photographing device 16, and an image analysis unit 18.

[0180] The light-emitting device 10C includes a plurality of light sources 101, similarly to the light-emitting device 10A of Specific Example 1. The plurality of light sources 101 are arranged on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. The distance L between the imaginary plane 82 and the collimating lens 12 is the same as the focal length f of the collimating lens 12.

[0181] The plurality of light sources 101 are arranged in the circumferential direction. Specifically, the plurality of light sources 101 include one light source 101a, four light sources 101b, twelve light sources 101c, and sixteen light sources 101d. The light source 101a is located on the optical axis 80 of the collimating lens 12. The four light sources 101b are arranged at equal intervals along a circle with a radius of r1 from the optical axis 80. The twelve light sources 101c are arranged at equal intervals along a circle with a radius of r2 (> r1) from the optical axis 80. The sixteen light sources 101d are arranged at equal intervals along a circle with a radius of r3 (> r2) from the optical axis 80.

[0182] Figure 21 This is a diagram showing examples of four lighting states of the light-emitting device 10C. The light-emitting device 10C sequentially switches Figure 21 to the first lighting state shown in (a) of Figure 21 the second lighting state shown in (b) of Figure 21 the third lighting state shown in (c) of Figure 21 and the fourth lighting state shown in (d) of . The first lighting state is a state in which the sixteen light sources 101d are simultaneously lit. The second lighting state is a state in which the twelve light sources 101c are simultaneously lit. The third lighting state is a state in which the four light sources 101b are simultaneously lit. The fourth lighting state is a state in which the light source 101a is lit.

[0183] Among the four photographed images respectively obtained by photographing in the first to fourth lighting states, the lighting state in which the brightness of each pixel becomes the maximum depends on the angle θ formed by the normal direction 86 of the surface of the object 2 imaged in the corresponding pixel and the optical axis 84 of the photographing device 16 (refer to Figure 20 ). Therefore, by confirming the normal image 20 generated from the four photographed images, the distribution of the angle θ can be grasped.

[0184] In Figure 2 and Figure 20In the example shown, a plurality of light sources 101 are arranged on an imaginary plane 82 perpendicular to the optical axis 80 of the collimating lens 12. In this case, due to the image surface curvature of the collimating lens 12, the light rays are deformed, which may affect the analysis of the captured image. Therefore, a plurality of light sources 101 may also be arranged on a spherical surface corresponding to the image surface curvature of the collimating lens 12.

[0185] Figure 22 FIG. is a diagram showing the structure of the inspection system 1D of Modification 2. As Figure 22 shown, the inspection system 1D includes a light emitting device 10D, a collimating lens 12, a half mirror 14, a photographing device 16, and an image analysis unit 18.

[0186] The light emitting device 10D includes a plurality of light sources 101 in the same manner as the light emitting device 10C of Modification 1. However, the plurality of light sources 101 are arranged on a spherical imaginary surface 88 having a center on the optical axis 80 of the collimating lens 12. The distance L1 between the light source 101 located on the optical axis 80 and the collimating lens 12 is the same as the focal length f of the collimating lens 12. The center and radius of the sphere having the imaginary surface 88 as the surface are set according to the image surface curvature of the collimating lens 12. Thereby, the influence of the light ray deformation caused by the image surface curvature of the collimating lens 12 can be suppressed.

[0187] In Figure 2 and Figure 20 shown in the example, the distance L between the light emitting devices 10A and 10C and the collimating lens 12 is set to be the same as the focal length f of the collimating lens 12. However, the distance L may also be set to be larger than the focal length f.

[0188] Figure 23 FIG. is a diagram showing the optical path when the distance L between the light emitting device 10A and the collimating lens 12 is set to be longer than the focal length f of the collimating lens 12. Figure 23 The virtual image 96 of the collimating lens 12, the virtual image 802 of the light emitting device 10A, and the virtual image 801 of the photographing device 16 are also shown. The collimating lens 12 and the virtual image 96 are symmetric with respect to the half mirror 14. The light emitting device 10 and the virtual image 802 are symmetric with respect to the half mirror 14. The photographing device 16 and the virtual image 801 are symmetric with respect to the plane (inspection surface) of the inspection object region 3 including the object 2.

[0189] As Figure 23 shown, the distance L between the light emitting device 10A and the collimating lens 12 is adjusted so that the irradiation solid angle ω at each point in the inspection object region 3 is the same as the observation solid angle Ψ of the photographing device 16. Thereby, even if the distance between the photographing device 16 and the object 2 is changed (that is, even if the photographing device 16 is brought closer to the object 2), at all points between the end points P and Q in the inspection object region 3, the state where the light emitting position where the brightness becomes the peak is the same is maintained. As a result, the error in the normal images 20X and 20Y can be reduced.

[0190] §3 Addendum

[0191] As described above, this embodiment includes the following disclosures.

[0192] (Structure 1)

[0193] An inspection system (1, 1A to 1D) that inspects the surface of an object to be inspected (2), wherein the inspection system (1, 1A to 1D) includes:

[0194] A light emitting device (10, 10A to 10D) that illuminates the object (2);

[0195] A collimating lens (12) disposed between the light emitting device (10, 10A to 10D) and the object (2); and

[0196] An imaging device (16) that images the object (2),

[0197] The light emitting device (10, 10A to 10D) varies the light emitting position,

[0198] The inspection system (1, 1A to 1D) further includes an image analysis unit (18) that generates a first analysis image (20, 20X, 20Y) in which the value of each pixel corresponds to the normal direction of the surface of the object (2) imaged in the corresponding pixel by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light emitting positions are different from each other.

[0199] (Structure 2)

[0200] In the inspection system (1, 1A to 1D) described in Structure 1,

[0201] The value of each pixel of the first analysis image (20, 20X, 20Y) represents the phase of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emitting position.

[0202] (Structure 3)

[0203] In the inspection system (1, 1A to 1D) described in Structure 1,

[0204] The value of each pixel of the first analysis image (20, 20X, 20Y) represents the light emitting position at which the brightness becomes a peak in the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emitting position.

[0205] (Structure 4)

[0206] In the inspection system (1, 1A to 1D) described in Structure 1,

[0207] The image analysis unit (18) further generates a second analysis image (22X, 22Y) by analyzing the plurality of captured images.

[0208] The value of each pixel of the second analysis image (22X, 22Y) is the amplitude of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light emission position.

[0209] (Structure 5)

[0210] In the inspection system (1, 1A, 1C, 1D) described in any one of Structures 1 to 4,

[0211] The light emitting device (10, 10A, 10C, 10D) includes a plurality of light sources (101, 101a to 101d), and sequentially switches the light source to be emitted among the plurality of light sources (101, 101a to 101d).

[0212] The plurality of light sources (101, 101a to 101d) are arranged on a plane (82) perpendicular to the optical axis (80) of the collimating lens (12), or on a spherical surface (88) having a center on the optical axis (80).

[0213] (Structure 6)

[0214] In the inspection system (1, 1A to 1D) described in any one of Structures 1 to 5,

[0215] The image analysis unit (18) further generates a composite image (24) obtained by synthesizing the plurality of captured images.

[0216] (Structure 7)

[0217] In the inspection system (1, 1A to 1D) described in any one of Structures 1 to 6,

[0218] The distance between the light emission position and the collimating lens (12) is equal to or greater than the focal length of the collimating lens (12).

[0219] (Structure 8)

[0220] In the inspection system (1, 1A to 1D) described in any one of Structures 1 to 7,

[0221] The inspection system (1, 1A to 1D) further includes a half mirror (14) disposed between the collimating lens (12) and the object (2).

[0222] The light emitted from the light-emitting device (10, 10A to 10D) and transmitted through the collimating lens (12) is reflected by the half mirror (14) and irradiates the object (2).

[0223] Based on the reflected light of the object (2), it passes through the half mirror (14) and enters the imaging device (16).

[0224] (Structure 9)

[0225] An inspection system (1, 1A to 1D) for inspecting the surface of an object (2), wherein the inspection system (1, 1A to 1D) includes:

[0226] A light-emitting device (10, 10A to 10D) for illuminating the object (2);

[0227] A collimating lens (12) disposed between the light-emitting device (10, 10A to 10D) and the object (2); and

[0228] An imaging device (16) for imaging the object,

[0229] The light-emitting device (10, 10A to 10D) varies the light-emitting position.

[0230] The inspection system (1, 1A to 1D) further includes an image analysis unit (18), which generates an analysis image by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other.

[0231] The value of each pixel of the analysis image is a feature quantity of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light-emitting position.

[0232] (Structure 10)

[0233] An inspection method for inspecting the surface of an object (2), wherein

[0234] The inspection method includes the following steps: while irradiating the object (2) with the light emitted from the light-emitting device (10, 10A to 10D) and transmitted through the collimating lens (12), imaging the object (2).

[0235] The step of performing the imaging includes the following steps:

[0236] Switching the light-emitting position in the light-emitting device (10, 10A to 10D); and

[0237] Obtaining a plurality of captured images from multiple captures in states where the light-emitting positions are different from each other.

[0238] The inspection method further includes the following steps: by analyzing the plurality of captured images, an analysis image (20, 20X, 20Y) is generated in which the value of each pixel corresponds to the normal direction of the surface of the object (2) imaged in the corresponding pixel.

[0239] The embodiments of the present invention have been described, but it should be considered that the embodiments disclosed this time are illustrative in all aspects rather than restrictive. The scope of the present invention is represented by the claims, and is intended to include meanings equivalent to the claims and all modifications within the scope.

[0240] Reference Signs

[0241] 1, 1A to 1D: inspection system; 2: object; 3: inspection target area; 5: display device; 6: memory card; 7: input device; 10, 10A to 10D: light-emitting device; 12: collimating lens; 14: half mirror; 16: imaging device; 18: image analysis unit; 20, 20X, 20Y: normal image; 21X, 21Y: differential image; 22X, 22Y: direct reflection image; 23X, 23Y: image; 24: composite image; 25: uneven image; 26: binary image; 28: difference image; 50, 51, 52: pixel; 60, 62: frame line; 80, 84: optical axis; 82: imaginary plane; 86: normal direction; 88: imaginary surface; 90, 92: light beam; 94: area; 96, 801, 802: virtual image; 101, 101a to 101d, 102: light source; 103: XY stage; 180: processor; 181: RAM; 182: display controller; 183: system controller; 184: I / O controller; 185: hard disk; 186: camera interface; 187: input interface; 189: communication interface; 190: memory card interface; 191: analysis program; F, F1: defect; L, L1: distance; LA, LB, LC: light; P, Q: end point; PA, PB, PC: light-emitting position.

Claims

1. An inspection system that inspects the surface of an object to be inspected, wherein, The inspection system includes: A light-emitting device for illuminating the object; A collimating lens disposed between the light-emitting device and the object; and An imaging device for imaging the object, The light-emitting device varies the light-emitting position, The inspection system further includes an image analysis unit that generates a first analysis image in which the value of each pixel corresponds to the normal direction of the surface of the object imaged in the corresponding pixel by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other. The value of each pixel of the first analysis image represents the phase of the following waveform or the light-emitting position at which the brightness peaks in the waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light-emitting position. The image analysis unit further generates a second analysis image by analyzing the plurality of captured images. The value of each pixel of the second analysis image is the amplitude of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light-emitting position.

2. The inspection system according to claim 1, wherein The light-emitting device includes a plurality of light sources, and the light source to emit light among the plurality of light sources is sequentially switched. The plurality of light sources are arranged on a plane perpendicular to the optical axis of the collimating lens or on a spherical surface having a center on the optical axis.

3. The inspection system according to claim 1, wherein The image analysis unit further generates a composite image obtained by synthesizing the plurality of captured images.

4. The inspection system according to claim 1, wherein The distance between the light-emitting position and the collimating lens is equal to or greater than the focal length of the collimating lens.

5. The inspection system according to claim 1, wherein The inspection system further includes a half mirror disposed between the collimating lens and the object. The light irradiated from the light-emitting device and transmitted through the collimating lens is reflected by the half mirror and irradiates the object. Based on the reflected light of the object, it passes through the half mirror and enters the imaging device.

6. An inspection system that inspects the surface of an object to be inspected, wherein, The inspection system includes: A light-emitting device for illuminating the object; A collimating lens disposed between the light-emitting device and the object; and An imaging device for imaging the object, The light-emitting device varies the light-emitting position, The inspection system further includes an image analysis unit that generates an analysis image by analyzing a plurality of captured images respectively obtained from multiple captures in states where the light-emitting positions are different from each other. The value of each pixel of the analysis image is the phase or amplitude of the following waveform, and the waveform represents the relationship between the brightness of the corresponding pixel in the plurality of captured images and the light-emitting position. The inspection system inspects for defects that affect the normal direction of the surface or defects that affect the degree of regular reflection of light at the surface based on the analysis image. The image analysis unit further generates a second analysis image by analyzing the plurality of captured images. The value of each pixel of the second analysis image is the amplitude of the following waveform, and the waveform represents the relationship between the luminance of the corresponding pixel in the plurality of captured images and the light-emitting position.

7. An inspection method for inspecting the surface of an object, wherein, the inspection method includes the following steps: irradiating the object with the light irradiated from the light-emitting device and transmitted through the collimator lens while capturing the object, the step of performing the capturing includes the following steps: switching the light-emitting position in the light-emitting device; and obtaining a plurality of captured images from multiple captures in a state where the light-emitting positions are different from each other, the inspection method further includes the following steps: generating an analysis image in which the value of each pixel corresponds to the normal direction of the surface of the object imaged in the corresponding pixel by analyzing the plurality of captured images, the value of each pixel of the analysis image represents the phase of the following waveform or the light-emitting position at which the luminance peaks in the waveform, and the waveform represents the relationship between the luminance of the corresponding pixel in the plurality of captured images and the light-emitting position, in the step of generating the analysis image, a second analysis image is further generated by analyzing the plurality of captured images, the value of each pixel of the second analysis image is the amplitude of the following waveform, and the waveform represents the relationship between the luminance of the corresponding pixel in the plurality of captured images and the light-emitting position.

Images