Optical inspection program, optical inspection system, processing device for optical inspection system, and optical inspection method

Abstract

To provide an optical inspection program that removes the image (wavelength selection pattern) of the wavelength selection area 26 included in the image when acquiring BRDF. [Solution] According to the embodiment, the optical inspection program causes the processor to remove the image of the wavelength selection unit from the image of the object surface captured through a wavelength selection unit that converts incident light into at least two different wavelength spectra, thereby generating a wavelength selection unit-removed image.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to an optical inspection program, an optical inspection system, a processing device for an optical inspection system, and an optical inspection method. [Background technology]

[0002] Non-contact surface inspection of objects is becoming increasingly important in various industries. Conventional methods involve illuminating an object by spectrally separating light rays, acquiring images of each spectrally separated ray using an image sensor, and estimating the direction of each ray to obtain information about the object's surface. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] U.S. Patent No. 5,675,407 [Non-patent literature]

[0004] [Non-Patent Document 1] W.L. Hows, “Rainbow schlieren and its application,” Applied Optics, vol. 23, No. 14, 1984. [Non-Patent Document 2] Hiroshi Ohno & Takahiro Kamikawa, “One-shot BRDF imaging system to obtain surface properties,” Optical Review volume 28, pages 655-661 (2021). [Overview of the project] [Problems that the invention aims to solve]

[0005] The problem that the present invention aims to solve is to provide an optical inspection program, an optical inspection system, a processing device for the optical inspection system, and an optical inspection method for removing images of wavelength-selective regions included in an image during BRDF acquisition. [Means for solving the problem]

[0006] According to one embodiment, the optical inspection program causes a processor to capture light incident from the surface of an object into at least two different wavelength spectra in a wavelength selection unit having at least two or more wavelength selection regions having a pattern, and then remove the image of the pattern of the wavelength selection unit from the captured image of the object surface onto which the image of the pattern of the wavelength selection unit is projected, thereby generating an image of the object surface which is a wavelength selection unit removed image. [Brief explanation of the drawing]

[0007] [Figure 1] A schematic diagram showing an optical inspection system according to the first embodiment. [Figure 2] A schematic block diagram of the processing unit for an optical inspection system. [Figure 3] A schematic flowchart of the processing performed by the processing unit of the optical inspection system shown in Figure 1. [Figure 4] A schematic diagram showing an optical inspection system according to the second embodiment. [Figure 5] A schematic flowchart of the processing performed by the processing unit of the optical inspection system shown in Figure 4. [Figure 6] Figure 4 shows the acquired image, the blue channel image, and the red channel image separated from the acquired image, all obtained using the optical device of the optical inspection system shown in Figure 4. [Figure 7] Figure 6 shows the frequency-space image obtained when the red channel image is transformed into Fourier space (frequency space) by a two-dimensional Fourier transform, as well as an image showing the masking process. [Figure 8] Wavelength-selective part removal image in real space obtained by inverse Fourier transform of the frequency-space image after masking. [Figure 9]It is a schematic flowchart of processing performed by the processing device of the optical inspection system according to the third embodiment. [Figure 10] It is a graph of the blue channel when the vertical axis represents blue channel intensity IB and the horizontal axis represents kx0*x. [Figure 11] It is a graph of the red channel when the vertical axis represents red channel intensity IR and the horizontal axis represents kx0*x. [Figure 12] It is a diagram showing a spatial frequency image obtained by Fourier transforming an image in which the wavelength selection unit is reflected. [Figure 13] It is a schematic diagram showing the optical inspection system according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Each embodiment will be described below with reference to the drawings. The drawings are schematic or conceptual, and the relationship between the thickness and width of each part, the size ratio between parts, and the like are not necessarily the same as those in reality. Furthermore, even when the same part is illustrated, the dimensions and ratios may be different depending on the drawing. In the present specification and each drawing of the present application, the same reference numerals are given to the same elements as those described above with respect to the already shown drawings, and detailed description thereof will be omitted as appropriate.

[0009] (First Embodiment) Hereinafter, the optical inspection system 10 according to the present embodiment will be described with reference to FIGS. 1 to 3.

[0010] In the present specification, light is a type of electromagnetic wave, and is defined to include X-rays, ultraviolet rays, visible light, infrared rays, microwaves, and the like. In other words, any electromagnetic wave that can be described by Maxwell's equations is acceptable. In the present embodiment, the light is assumed to be visible light, for example, having a wavelength in the range of 400 nm to 750 nm.

[0011] FIG. 1 is a schematic cross-sectional view of the optical inspection system 10 according to the present embodiment.

[0012] The optical inspection system 10 according to the present embodiment includes an optical device 12 and a processing device 14.

[0013] The optical device 12 includes an imaging unit 24 and a wavelength selection unit (multi-wavelength aperture) 26.

[0014] The imaging unit 24 is directed towards the part of the surface of object S that is to be illuminated. The imaging unit 24 has an imaging optical element 42 and an image sensor (color image sensor) 44. The imaging optical element 42 is, for example, an imaging lens. The imaging optical element 42 has a focal length of f. In Figure 1, the imaging lens is schematically represented by a single lens, but it may be a lens set composed of multiple lenses. Alternatively, the imaging optical element 42 may be a concave mirror, a convex mirror, or a combination thereof. In other words, the imaging optical element 42 can be any optical element that has the function of focusing a group of light rays emitted from a point on the object S, i.e., the object point, to its conjugate image point on the image sensor 44. The process by which the imaging optical element 42 focuses (concentrates) a group of light rays emitted from the object point on the surface of object S to the image point is called imaging. Alternatively, it is said that the object point is moved to the image point (the conjugate point of the object point). In this way, the object point and the image point are associated as conjugates through the imaging optical element 42. The plane of conjugate points to which a group of light rays emitted from a sufficiently distant object point are moved by the imaging optical element 42 is called the focal plane of the imaging optical element 42. Furthermore, the line perpendicular to the focal plane and passing through the center of the imaging optical element 42 is called the optical axis L. The point where the optical axis L intersects the focal plane is called the focus.

[0015] Furthermore, an xyz Cartesian coordinate system is defined, where the z-axis is the direction along the optical axis L, and the x-axis is perpendicular to the z-axis, and the y-axis is perpendicular to both the x-axis and the z-axis. Here, the xyz coordinate system is defined for the wavelength selection unit 26, and the origin O is set on the boundary between the first wavelength selection region 52 and the third wavelength selection region 54. The z-axis intersects the wavelength selection unit 26. The x-axis intersects the multiple wavelength selection regions 52 and 54. The y-axis follows the multiple wavelength selection regions 52, 54, and 56. However, this is not limited to the multiple wavelength selection regions 52, 54, and 56; the y-axis may also intersect with wavelength selection regions other than the multiple wavelength selection regions 52, 54, and 56. In other words, the wavelength selection unit 26 only needs to have at least two or more wavelength selection regions intersecting the x-axis.

[0016] The image sensor 44 can be anything as long as it has at least two pixels arranged in a row that convert light into electrical signals. For example, it can be an area sensor with pixels arranged in an area, or a line sensor with pixels arranged in a line.

[0017] The pixels of the image sensor 44 are assumed to have at least two different color channels and be able to distinguish at least two different wavelengths. That is, they are able to distinguish between a first wavelength and a second wavelength. For example, the first wavelength is 450 nm blue light and the second wavelength is 650 nm red light. However, the wavelengths are not limited to these and can be any. Here, the color channel sensitive to the first wavelength is called the blue channel, and the color channel sensitive to the second wavelength is called the red channel. The image sensor 44 according to this embodiment may also have three color channels, R, G, and B, in each pixel. For this reason, in this embodiment, it is preferable that each pixel can further receive green G light with a wavelength of 550 nm in its own independent color channel.

[0018] The wavelength selection unit 26 in this embodiment has a stripe shape parallel to the y-axis. The wavelength selection unit 26 is provided between the surface of the object S and the imaging unit 24.

[0019] The wavelength selection unit 26 has at least two or more wavelength selection regions 52, 54. Two of these wavelength selection regions are designated as the first wavelength selection region 52 and the second wavelength selection region 54.

[0020] In this embodiment, the first wavelength selection region 52 and the second wavelength selection region 54 each extend, for example, along the y-axis. The first wavelength selection region 52 and the second wavelength selection region 54 intersect with the x-axis.

[0021] The first wavelength-selective region 52 allows light rays having a wavelength spectrum including the first wavelength to pass through. Therefore, in this embodiment, the first wavelength-selective region 52 transmits light rays of the first wavelength. Here, "allowing light rays to pass through" means directing the light rays from the object point to the image point by transmission or reflection. On the other hand, the first wavelength-selective region 52 substantially shields light rays of the second wavelength. Here, "shielding" means not allowing the light rays to pass through. In other words, it means not directing the light rays from the object point to the image point.

[0022] The second wavelength-selective region 54 allows light rays of the second wavelength to pass through. Therefore, in this embodiment, the second wavelength-selective region 54 transmits light rays of the second wavelength. On the other hand, the second wavelength-selective region 54 substantially blocks light rays of the first wavelength.

[0023] The wavelength selection unit 26 has a pattern because it comprises at least two or more wavelength selection regions 52, 54. This is called the wavelength selection unit pattern. Alternatively, it may simply be called the wavelength selection unit 26. When the surface of object S is imaged by the imaging unit 24 through the wavelength selection unit 26, the wavelength selection unit pattern (or wavelength selection unit 26) is projected and appears in the image.

[0024] The processing unit 14 includes, for example, a processor 61 (control unit), a ROM (storage unit) 62, a RAM 63, an auxiliary storage device 64 (storage unit), a communication interface 65 (communication unit), and an input unit 66.

[0025] The processor 61 is the central part of the computer that performs calculations and control necessary for the processing of the processing unit 14, and comprehensively controls the entire processing unit 14. Based on programs such as system software, application software, or firmware stored in a storage unit such as ROM 62 or auxiliary storage device 64, the processor 61 executes control to realize various functions of the processing unit 14. The processor 61 includes, for example, a CPU (central processing unit), MPU (micro processing unit), DSP (digital signal processor), ASIC (Application Specific Integrated Circuit), or FPGA (Field Programmable Gate Array). Alternatively, the processor 61 is a combination of several of these. The processing unit 14 may have one processor 61 or multiple processors 61.

[0026] ROM62 corresponds to the main memory of a computer centered around the processor 61. ROM62 is a non-volatile memory used exclusively for reading data. ROM62 stores the program described above. Furthermore, ROM62 stores data or various settings used by the processor 61 in performing various processes.

[0027] RAM63 corresponds to the main memory of a computer centered around the processor 61. RAM63 is memory used for reading and writing data. RAM63 is used as a so-called work area, where data temporarily used by the processor 61 during various processes is stored.

[0028] The auxiliary storage device 64 corresponds to the auxiliary storage device of a computer centered on the processor 61. The auxiliary storage device 64 is, for example, an EEPROM (electrically erasable programmable read-only memory) (registered trademark), an HDD (hard disk drive), or an SSD (solid state drive). The auxiliary storage device 64 may also store the above-mentioned programs. In addition, the auxiliary storage device 64 stores data used by the processor 61 in performing various processes, data generated by processing by the processor 61, or various setting values.

[0029] The program stored in the ROM 62 or auxiliary storage device 64 includes a program for controlling the processing unit 14. For example, an optical inspection program is preferably stored in the ROM 62 or auxiliary storage device 64.

[0030] The communication interface 65 is an interface for communicating with other devices via a network or the like, either by wire or wireless connection, receiving various information transmitted from other devices, and transmitting various information to other devices. The processing unit 14 acquires image data obtained by the image sensor 44 via the communication interface 65.

[0031] The processing unit 14 preferably includes an input unit 66, such as a keyboard, for inputting the arrangement and type of the wavelength selection unit 26. The input unit 66 may also be capable of receiving various types of information wirelessly from the processor 61 via a communication interface 65.

[0032] The processing unit 14 performs various functions by causing the processor 61 to execute programs stored in the ROM 62 and / or auxiliary storage device 64, etc. It is also preferable that the control program of the processing unit 14 is not stored in the ROM 62 and / or auxiliary storage device 64 of the processing unit 14, but is instead located on an appropriate server or cloud. In this case, the control program is executed while communicating with the processor 61 of, for example, the optical inspection system 10 via the communication interface 65. That is, the processing unit 14 according to this embodiment may be part of the optical inspection system 10, or it may be located on a server or cloud of a system in various inspection facilities, separate from the optical inspection system. Therefore, it is also preferable that the optical inspection program resides on a server or cloud, rather than being stored in the ROM 62 or auxiliary storage device 64, and that the program is executed while communicating with the processor 61 of, for example, the optical inspection system 10 via the communication interface 65. Consequently, the processor 61 (processing unit 14) can execute the optical inspection program (optical inspection algorithm) described later.

[0033] The processor 61 (processing unit 14) controls the timing of image data acquisition by the image sensor 44, the acquisition of image data from the image sensor 44, and other related functions.

[0034] The basic operation of the optical inspection system 10 described above will now be explained.

[0035] Figure 3 shows a flowchart of the optical inspection algorithm (optical inspection program) executed by the processing unit 14 (processor 61) of the optical inspection system 10 of this embodiment.

[0036] The flow of the optical inspection algorithm in this embodiment is as follows: First, an image is acquired with the wavelength selection pattern projected (ST1). In this embodiment, the processing unit 14 controls the image sensor 44 and acquires an image of the light from the surface of the object S that has passed through the wavelength selection unit 26. The directional distribution of reflected light from a point (referred to here as an object point) on the surface of object S can be represented by a distribution function called the BRDF (Bidirectional Reflectance Distribution Function). Generally, the BRDF changes depending on the surface properties and shape. For example, if the surface is rough, the reflected light spreads in various directions, resulting in a broad BRDF distribution. In other words, reflected light exists over a wide range of angles. On the other hand, if the surface is mirror-like, the reflected light consists almost entirely of specular reflection, resulting in a narrow BRDF distribution. Thus, the BRDF reflects the surface properties and shape of the object's surface. Here, surface properties and shape can refer to surface roughness, micron-sized irregularities, surface inclination, strain, etc. In short, anything related to the height distribution of the surface is acceptable. When the surface properties and shape consist of a fine structure, the typical structural scale can be nanoscale, micronscale, milliscale, or any other scale. Therefore, the image acquired by the image sensor 44 includes images of object points O1 and O2 on the surface of object S. Furthermore, the captured image also includes an image of the wavelength selection unit 26 positioned between the imaging unit 24 and the surface of object S. Consequently, the captured image includes the images of object points O1 and O2 on the surface of object S, colored with the wavelength selection regions 52 and 54. In this image, the boundaries of the wavelength selection regions 52 and 54 are, for example, visible.

[0037] According to this embodiment, the wavelength spectrum of light acquired by the pixels of the image sensor 44 of the imaging unit 24 changes depending on the BRDF at object points O1 and O2. For example, as shown in Figure 2, if the surface of object S at the first object point O1 is flat, the BRDF will have a narrow distribution like the first BRDF. The light reflected at the first object point O1 will pass through the first wavelength selection region 52 of the wavelength selection unit 26. In other words, it will not pass through the second wavelength selection region 54.

[0038] On the other hand, if a minute defect exists on the surface of object S at the second object point O2, the BRDF will have a wider distribution than the first BRDF, as is the case with the second BRDF. The light reflected at the second object point O2 will pass through both the first wavelength selection region 52 and the second wavelength selection region 54. Alternatively, some light will not pass through either of the wavelength selection regions 52 or 54 of the wavelength selection unit 26 and will deviate from the imaging unit 24. If a large amount of light deviates from the imaging unit 24, the intensity of light at the image point corresponding to the second object point O2 will decrease. As a result, the wavelength spectra of the light arriving at the respective image points of the first object point O1 and the second object point O2 will be different. Different wavelength spectra will result in different colors being acquired by the pixels of the image sensor. This has the effect of allowing the difference in BRDF to be identified by color. Furthermore, being able to identify the difference in BRDF has the effect of allowing the presence or absence of minute defects at object points O1 and O2 to be determined. It is generally known that if the surface properties and shape of an object S differ, the BRDF (Body-Resistant Fiber Deposition) will also differ. Therefore, being able to distinguish between differences in BRDF has the effect of being able to distinguish between the surface properties and shape of object S.

[0039] The fact that pixel colors differ depending on the BRDF means that when viewing an image of any color channel, the pixel value changes depending on the BRDF. In other words, the image of any color channel will exhibit brightness and darkness that reflect the difference in BRDF. This has the effect of allowing us to identify the difference in BRDF by identifying the brightness and darkness of the image in each color channel. If we can identify the difference in BRDF, we can identify the surface properties and shape of object S.

[0040] The fact that the wavelength selection regions 52 and 54 that pass through differ depending on the BRDF means that when viewing images of at least two color channels, the pixel values ​​of each color channel at corresponding image points I1 and I2 change in a correlated manner depending on the BRDF difference. For example, in the pixel where the first BRDF of the first object point O1 is imaged, the pixel value of the blue channel increases, and the pixel value of the red channel becomes almost 0. In contrast, in the pixel where the second BRDF of the second object point O2 is imaged, the pixel value of the blue channel decreases, and the pixel value of the red channel increases in a correlated manner. In other words, the pixel values ​​of each color channel change in a correlated manner depending on the properties of the object point. This correlation has the effect of allowing the difference in BRDF to be identified. If the difference in BRDF can be identified, the properties and shape of the surface of object S can be identified.

[0041] However, the color changes caused by the wavelength selection pattern 26 reflected in the image may be indistinguishable from the color changes caused by the BRDF. In other words, in the image of any color channel, the wavelength selection pattern changes the pixel values. This results in brightness and darkness in each color channel. Alternatively, the image of at least two color channels changes in a correlated manner due to the wavelength selection pattern.

[0042] Next, an image with the wavelength-selective pattern removed (wavelength-selective-removed image) is generated (ST2). To remove the wavelength selection pattern from the acquired image, for example, a surface of an object S with a uniform BRDF can be prepared in advance and captured by the imaging unit 24 through the wavelength selection unit 26. This image can then be stored as a reference image in, for example, the ROM 62 and / or auxiliary storage device 64, and the reference image can be subtracted from the acquired image for each color channel. The surface of the reference object S can be anything, such as a reflective, monochromatic surface or a matte, monochromatic surface. This makes it possible to remove the wavelength selection pattern from the acquired image.

[0043] Finally, the difference in BRDF of each object point O1 and O2 is identified using that image (the image from which the wavelength-selective pattern has been removed) (ST3). In other words, information about the surface of object S is obtained based on the image from which the wavelength-selective pattern has been removed. The wavelength-selective removal image has the color changes caused by the wavelength-selective pattern removed. On the other hand, the information on the color changes caused by the BRDF at each object point O1, O2 on the surface of object S remains. In other words, the processing device 14 can identify the BRDF of each object point O1, O2 by identifying the color change information. By removing the wavelength selection pattern from the acquired image, the processing device 14 can identify differences in BRDF based on changes in each color channel without misinterpreting them as being due to the wavelength selection pattern. Furthermore, if differences in BRDF can be identified, the properties and shape of the surface of object S can be identified.

[0044] Thus, according to this embodiment, it is possible to provide an optical inspection program, an optical inspection system 10, a processing device 14 for the optical inspection system 10, and an optical inspection method that remove the image of the wavelength selection unit 26 (wavelength selection unit pattern) included in the image when acquiring BRDF. Furthermore, according to this embodiment, by performing a process to remove the wavelength selection unit pattern from the acquired image, it is possible to identify the BRDF by the changes in each color channel without misrecognizing them as being due to the wavelength selection unit pattern.

[0045] (Second Embodiment) The optical inspection system 10 according to the second embodiment will be described with reference to Figures 4 to 8. This embodiment is a modified example of the optical inspection system 10 according to the first embodiment, and the same reference numerals are used as much as possible for components that are the same as or have the same function as those described in the first embodiment, and detailed explanations are omitted.

[0046] Figure 4 shows a schematic cross-sectional view in the xz plane of the optical device 12 of the optical inspection system 10 according to this embodiment. Here, the optical axis L of the imaging optical element 42 of the imaging unit 24 of the optical device 12 is defined as the z axis. The axes perpendicular to it are defined as the x axis and the y axis.

[0047] The basic configuration of the optical device 12 of the optical inspection system 10 in this embodiment is essentially the same as that of the optical device 12 of the optical inspection system 10 in the first embodiment.

[0048] The optical device 12 of this embodiment further includes an illumination unit 22 and a beam splitter 28 in addition to the optical device 12 described in the first embodiment.

[0049] The illumination unit 22, although not shown in the figure, consists of a light source, an aperture, an illumination lens, etc. The light source can be anything that emits light. Here, the light source is, for example, an LED that emits white light. The light source's ON / OFF emission is controlled by the processing device 14. The aperture is a light-shielding plate with a slit. The light source is positioned at the focal plane of the illumination lens. In this configuration, the light emitted from the LED is partially shielded by the aperture and partially passes through. In the cross-sectional view shown in Figure 4, the light emitted from the illumination unit 22 is substantially parallel light. However, the parallel light may have a small divergence angle of a few degrees. Also, when this light is projected onto a cross-section other than this cross-section, it does not necessarily have to be parallel light; it may be divergent light.

[0050] Parallel light from the illumination unit 22 is irradiated onto the surface of object S via the beam splitter 28. The beam splitter 28 may be a polarizing beam splitter or an unpolarized beam splitter. When a polarizing beam splitter is used, the light scattered from the surface of object S undergoes a rotation in polarization, is transmitted, and enters the imaging unit 24. In other words, this has the effect of extracting only the scattered light.

[0051] In this embodiment, the wavelength selection unit 26 has three wavelength selection regions 52, 54, and 56. These three wavelength selection regions are designated as the first wavelength selection region 52, the second wavelength selection region 54, and the third wavelength selection region 56.

[0052] The first wavelength-selective region 52 allows light rays having a wavelength spectrum including the first wavelength to pass through. On the other hand, the first wavelength-selective region 52 substantially blocks light rays of the second and third wavelengths. However, it is not necessary to completely block the second and third wavelengths; for example, it is sufficient to reduce the intensity of those light rays to half or less.

[0053] The second wavelength-selective region 54 allows the wavelength spectrum containing light rays of the second wavelength to pass through. On the other hand, the second wavelength-selective region 54 substantially blocks light rays of the first and third wavelengths. However, it is not necessary to completely block the first and third wavelengths; for example, it is sufficient to reduce the intensity of their light to half or less.

[0054] The third wavelength-selective region 56 allows the wavelength spectrum containing the third wavelength of light to pass through. On the other hand, the third wavelength-selective region 56 substantially blocks the first wavelength of light and the second wavelength of light. However, it is not necessary to completely block the first and second wavelengths; for example, it is sufficient to reduce the intensity of their respective light to half or less.

[0055] In this embodiment, for example, the first wavelength is 450 nm blue light, the second wavelength is 650 nm red light, and the third wavelength is 550 nm green light. However, the wavelengths are not limited to these and can be any.

[0056] The image sensor 44 may be an area sensor or a line sensor. In this embodiment, the image sensor 44 is an area sensor, and each pixel has three color channels: red (R), blue (B), and green (G). In other words, the image sensor 44 can substantially receive blue light with a wavelength of 450 nm, red light with a wavelength of 650 nm, and green light with a wavelength of 550 nm, each in an independent color channel. However, each color channel does not need to be completely independent, and each color channel may have a slight sensitivity to wavelengths other than its corresponding wavelength.

[0057] In this embodiment, in the xyz coordinate system, the second wavelength selection region 54 is located at the origin O. The first wavelength selection region 52 is located adjacent to the second wavelength selection region 54 in the -x-axis direction. The third wavelength selection region 56 is located adjacent to the second wavelength selection region 54 in the +x-axis direction. The first wavelength selection region 52, the second wavelength selection region 54, and the third wavelength selection region 56 are assumed to extend along the y-axis.

[0058] In the processing unit 14 of the optical inspection system 10 of this embodiment, the optical inspection algorithm (program) executed by the processing unit 14 described in the first embodiment is operated. The light rays are schematically shown as projected onto this cross-section.

[0059] The operation of the optical inspection system 10 of this embodiment will now be described.

[0060] In the cross-section shown in Figure 4, the processing device 14 emits light from the light source of the illumination unit 22, irradiating the surface of object S with parallel light from the illumination unit 22. In this case, if the surface of object S is a mirror surface, the parallel light is reflected. At this time, the first BRDF has a very narrow light distribution because it is parallel light. In other words, it can have a narrower light distribution compared to when the illumination is diffuse light instead of parallel light. On the other hand, the second BRDF has a wide light distribution due to minute defects on the surface of object S. This is the same whether the illumination is parallel light or diffuse light. As a result, by irradiating with parallel light from the illumination unit 22, a large difference can be made in the degree of light distribution spread between the first BRDF and the second BRDF. This has the effect of making a large difference in the wavelength spectra of the two when light passes through the wavelength selection unit 26. In other words, it has the effect of increasing the discrimination sensitivity of the BRDF and improving inspection accuracy.

[0061] Light that can be represented by the first BRDF passes through the first wavelength selection region 52, the second wavelength selection region 54, or the first wavelength selection region 52 and the second wavelength selection region 54. In this embodiment, at the first object point O1, the light that can be represented by the first BRDF passes through the first wavelength selection region 52 and the second wavelength selection region 54. On the other hand, light that can be represented by the second BRDF passes through, for example, the first wavelength selection region 52, the second wavelength selection region 54, and the third wavelength selection region 56.

[0062] Therefore, the wavelength spectra of light represented by the first BRDF and light represented by the second BRDF are different. Furthermore, these can be distinguished by pixels with three color channels. On the other hand, if a third wavelength selection region 56 does not exist in the wavelength selection unit 26, it becomes difficult to distinguish between the first BRDF and the second BRDF. This is because light represented by the first BRDF may pass near the boundary between the two adjacent wavelength selection regions 52 and 54.

[0063] Therefore, the effect is that such problems can only be solved if the wavelength selection unit 26 has three or more wavelength selection regions 52, 54, and 56. In other words, even if light passes near the boundary between two adjacent wavelength selection regions 52 and 54, if the wavelength selection unit 26 has three or more wavelength selection regions 52, 54, and 56, it can distinguish different BRDFs without mistakenly recognizing them as the same.

[0064] Figure 5 shows a flowchart of the optical inspection algorithm (optical inspection program) executed by the processing unit 14 (processor 61) of the optical inspection system 10 of this embodiment.

[0065] As described in the first embodiment, the flow of the optical inspection algorithm of this embodiment is to first acquire an image onto which the wavelength selection pattern is projected (ST1). The acquired image obtained by the image sensor 44 includes images of object points O1 and O2 on the surface of object S. The acquired image at this time also includes an image of the wavelength selection unit 26, which is positioned between the imaging unit 24 and the surface of object S. The acquired image has the wavelength selection pattern superimposed on the images of object points O1 and O2 on the surface of object S, as the color of the wavelength selection regions 52 and 54.

[0066] As an example, an image of a region containing minute defects on a white plastic plate was captured by the imaging unit 24 through the wavelength selection unit 26 using the optical device 12 shown in Figure 4. An example of the acquired image is shown in the left panel of Figure 6. In the image shown in Figure 6, the image was obtained with pixel values ​​that are strongly red on the upper side and strongly blue on the lower side, with the boundary near the y-axis in Figure 6.

[0067] Next, an image with the wavelength-selective pattern removed (wavelength-selective-removed image) is generated (ST2). In this embodiment, when generating an image from which the wavelength-selective pattern has been removed, the acquired image is Fourier transformed, and the wavelength-selective pattern is removed in Fourier space. In other words, the spatial frequency components corresponding to the wavelength-selective pattern are removed in Fourier space.

[0068] Thus, by removing the spatial frequency components corresponding to the wavelength selection pattern in Fourier space, the image can be reliably and accurately removed without depending on the brightness (magnitude of the average pixel value) of the acquired image. On the other hand, when applying a method that involves acquiring a reference image in advance and removing the image from it, it is necessary to match the brightness of the reference image and the acquired image. In this case, it is also necessary to appropriately set the brightness of the illumination during image acquisition.

[0069] As shown in the right-hand diagram of Figure 6, the processing unit 14 separated the blue channel image and the red channel image from the acquired image. The following section describes the red channel image. The same process is followed for the blue channel image.

[0070] The image in the red channel has two orthogonal axes, in the x and y directions. The wavelength selection area 26 captured in the red channel image is a rectangle with large, constant pixel values ​​in this image. Let this be region A, and let its width be D in the x direction.

[0071] The image in the red channel is transformed into Fourier space (frequency space) using a two-dimensional Fourier transform. The image after the Fourier transform is shown in Figure 7. This is defined as the frequency space image (the captured image of the red channel in frequency space). The frequency space image is k x and k y It has two orthogonal axes. The reflection image of the wavelength selection unit 26, that is, the principal component corresponding to region A, is located within a radius k0 from the center of the image. Here, k0 is determined using the region width D of region A, k0 = 2π / D (1) This can be expressed as follows. Therefore, the processing unit 14 sets the frequency-space image of the region within radius k0 to 0. Alternatively, it sets it to a constant value that is sufficiently small compared to the surrounding intensity (the size of the pixel values ​​in the frequency-space image). In other words, the processing unit 14 masks the region within radius k0 and generates a wavelength-selective area-removed image in the frequency space.

[0072] In the acquired image, the image of the wavelength selection unit 26 is obtained as a component of the region where the change in brightness value is gradual (low-frequency component), while scratches on the surface of object S are obtained as a component of the region where the change in brightness value is steep (high-frequency component). In general, images in this frequency space have low-frequency components near the center and high-frequency components as you move away from the center.

[0073] Therefore, by masking a region with radius k0 from the center of the frequency-space image, low-frequency components are cut out, effectively functioning like a high-pass filter (ST21).

[0074] The processing unit 14 then performs an inverse Fourier transform on the frequency-space image after masking (the wavelength-selective part-removed image in frequency space) to return it to real space. As a result, the processing unit 14 generates a wavelength-selective part-removed image in real space (see Figure 8) (ST2). As shown in Figure 8, the wavelength-selective part-removed image in real space can remove the image of the wavelength-selective part 26 from the acquired image obtained by the image sensor 44 shown in Figure 6. As is clear from Figure 8, the wavelength-selective part 26 that was captured in the image is removed, and minute defects are made clearer in the image.

[0075] In this embodiment, the processing device 14 processes the image of the red channel to obtain a wavelength-selective deselected image in real space. However, it is possible to perform the same processing on the image of the other color channel, namely the blue channel in this case, to obtain a similar wavelength-selective deselected image in real space.

[0076] Therefore, by performing the above-described processing on the red channel image and the blue channel image, the wavelength selection unit 26 reflected in both images is removed, and minute defects can be made clearer. By performing arithmetic operations such as adding and averaging or multiplying these images to create a correlated image, minute defects can be made even clearer than with a single image. This has the effect of enabling highly accurate inspection of the surface of object S.

[0077] As described in the first embodiment, finally, the difference in BRDF of each object point O1 and O2 can be identified using the image (the image obtained by removing the wavelength selection pattern from the acquired image) (ST3).

[0078] The optical device 12 according to this embodiment includes a support portion 72 that supports the outer edge of the wavelength selection portion 26, and a first adjustment portion 74 that allows the support portion 72 to be rotated around the axis of the optical axis L.

[0079] The first adjustment unit 74 preferably uses, for example, a servo motor and is controlled wirelessly or by wire by the processing unit 14 to control the arrangement of the support unit 72, that is, the arrangement of the wavelength selection unit 26 around the axis of the optical axis L (around the axis of the origin O). In this way, the first adjustment unit 74 can rotate the wavelength selection unit 26 to a desired angle, for example, around the axis of the optical axis L, relative to the imaging unit 24. If the BRDF has special anisotropy, the processing unit 14 can obtain a highly accurate BRDF distribution by rotating the wavelength selection unit 26 around the axis of the optical axis L using the first adjustment unit 74 while imaging the surface of the object S with the image sensor 44.

[0080] Such support portion 72 and first adjustment portion 74 can also be used in the wavelength selection portion 26 of the optical device 12 of the optical inspection system 10 described in the first embodiment.

[0081] According to this embodiment, an optical inspection program, an optical inspection system 10, a processing device 14 for the optical inspection system 10, and an optical inspection method can be provided that remove the image (wavelength selection pattern) of the wavelength selection unit 26 included in the image when acquiring BRDF.

[0082] (modified version) The wavelength-selective portion 26 projected onto the image may be significantly larger than the representative size of the minute defect to be inspected. In this case, removing the wavelength-selective portion 26 projected onto the image acquired by the image sensor 44 can also be achieved by applying a spatial high-pass filter to the captured image.

[0083] (Third embodiment) The algorithm of the optical inspection system 10 according to the third embodiment and the inspection apparatus using it will be described below with reference to Figures 9 to 12. This embodiment is a modified version of the optical inspection system 10 according to the first and second embodiments.

[0084] FIG. 9 shows a flowchart of an optical inspection algorithm (optical inspection program) executed by the processing device 14 (processor 61) of the optical inspection system 10 according to the present embodiment.

[0085] In the present embodiment, the wavelength selection unit pattern is set as a sine wave (ST11). However, it is assumed that the wavelength selection unit pattern changes in the x-direction and is uniform in the y-direction as shown in FIG. 4. Further, let the spatial frequency of the sine wave be k x0 . Here, the statement that the wavelength selection unit pattern is a sine wave means that, for example, the transmission intensity when blue light passes through the wavelength selection unit 26 (the overall transmission intensity (transmittance) of the wavelength spectrum corresponding to blue when the wavelength selection region is of a transmission type) changes in a sine wave shape along the x-direction. It also means that, at the same time, the transmission intensity when red light passes through (the overall transmission intensity (transmittance) of the wavelength spectrum corresponding to red when the wavelength selection region is of a transmission type) changes in a sine wave shape along the x-direction. However, the phase (peak position) and transmission intensity of the sine wave for blue light and the sine wave for red light may be different. Here, the phases are shifted by 90° and superimposed. As a result, a region where red light is strongly transmitted has weak blue transmission, and a region where blue light is strongly transmitted has weak red transmission.

[0086] When the surface of the object S is imaged through the wavelength selection unit pattern, the wavelength selection unit pattern is reflected in the image acquired by the image sensor 44 (ST1). As shown in FIG. 10, let the blue channel intensity be I B and, as shown in FIG. 11, let the red channel intensity be I R which are taken as the vertical axis of each graph. Further, the horizontal axis of each graph shown in FIG. 10 and FIG. 11 is k x0 *x. Here, k x0 is the spatial frequency of the sine wave.

[0087] Fourier transform is performed on the image in which the wavelength selection unit 26 is reflected, and a spatial frequency image is calculated. Then, the wavelength selection unit 26 can be expressed as two points on the k x axis as shown in FIG. 12. That is, +k x on the k x0 axis and -kx0 These are the two points.

[0088] k x +k on the axis x0 and, -k x0 By masking these two points, the wavelength selection area 26 that appears in the image can be removed (ST21). More specifically, k x +k on the axis x0 and, -k x0 The frequency-space image of the region between these two points is set to 0. Alternatively, it is set to a constant value that is sufficiently small compared to the surrounding intensity (the magnitude of the pixel values ​​in the frequency-space image).

[0089] Here, by making the wavelength selection unit 26 a sine wave, the area to be masked can be reduced to two points, making it smaller. This has the effect of making it possible to remove the wavelength selection unit 26 that is reflected in the image without losing as much information as possible about minute defects.

[0090] Then, an image with the wavelength-selective pattern removed (wavelength-selective-removed image) is generated. The processing unit 14 performs an inverse Fourier transform on the frequency-space image after masking (wavelength-selective-removed image in frequency space) to return it to real space. Therefore, the processing unit 14 generates a wavelength-selective-removed image in real space (ST2).

[0091] In this way, by optimizing the wavelength selection pattern of the wavelength selection unit 26, it is possible to remove the wavelength selection unit 26 reflected in the image while preserving the detailed information of the surface of the object S.

[0092] As described in the first embodiment, finally, the difference in BRDF of each object point O1 and O2 can be identified using the image (the image obtained by removing the wavelength selection pattern from the acquired image) (ST3).

[0093] According to this embodiment, an optical inspection program, an optical inspection system 10, a processing device 14 for the optical inspection system 10, and an optical inspection method can be provided that remove the image (wavelength selection pattern) of the wavelength selection unit 26 included in the image when acquiring BRDF.

[0094] In this embodiment, the wavelength selection unit 26 is a sine wave. However, it is not necessarily required to be a sine wave; if the wavelength selection unit 26 has a periodic pattern, the spatial frequency corresponding to that period becomes the principal component in the frequency space. Therefore, the area to be masked can be the neighborhood of two points. This has the effect of making it possible to remove the wavelength selection unit 26 that has appeared in the image without losing as much information about minute defects as possible.

[0095] (Fourth embodiment) (Line sensor type) The optical inspection system 10 according to this embodiment will be described below with reference to Figure 13. This embodiment is a modified version of the optical inspection system 10 according to the first, second, and third embodiments. Components that are the same as or have the same function as those described in the second and third embodiments are given the same reference numerals as much as possible, and detailed descriptions are omitted.

[0096] Figure 13 shows a perspective view of the optical device 12 of the optical inspection system 10 according to this embodiment. The optical device 12 projected onto the first cross-section S1 in Figure 13 is basically the same as the optical device 12 described in the second embodiment (see Figure 4).

[0097] First, in this embodiment, the optical system was constructed without using a beam splitter. This has the effect of reducing the number of optical elements required.

[0098] The image sensor 44 is a line sensor. However, it may also be an area sensor.

[0099] The first cross-section S1 is defined as a cross-section that includes the optical axis of the imaging optical element 42 and is perpendicular to the longitudinal direction of the line sensor 44. When light from the illumination unit 22 is projected onto the first cross-section S1, it is defined as parallel light. On the other hand, the second cross-section S2 is defined as a cross-section perpendicular to the first cross-section S1. When light from the illumination unit 22 is projected onto the second cross-section S2, it does not have to be parallel light; it may be diffuse light. Here, it is defined as diffuse light.

[0100] The wavelength selection unit 26 comprises multiple (for example, three) wavelength selection regions 52, 54, and 56. Each wavelength selection region 52, 54, and 56 intersects the x-axis and has a long stripe shape along the y-axis. In the first cross-section S1, the three wavelength selection regions 52, 54, and 56 are arranged. In other words, in this cross-section S1, the wavelength selection regions 52, 54, and 56 of the wavelength selection unit 26 change along the x-axis. On the other hand, in a cross-section parallel to the second cross-section S2 which is perpendicular to this cross-section S1, the wavelength selection regions of the wavelength selection unit 26 do not change.

[0101] The illumination unit 22 illuminates the surface of object S, forming an illumination field F. The illumination field F of the illumination unit 22 is formed in a linear or rectangular shape on the surface of object S. A first object point within the illumination field F is imaged onto a first image point on the line sensor 44 by the imaging optical element 42. At the first object point, the BRDF becomes the first BRDF. The first ray is contained within the first BRDF.

[0102] When a light ray is projected onto the first cross-section S1, the extent of the distribution of the first BRDF can be identified by the wavelength spectrum of the light that has passed through the wavelength selection regions 52, 54, and 56 of the wavelength selection unit 26. When the light reaches the image point I1 in the line sensor 44, the line sensor 44 is identified as a color corresponding to its wavelength spectrum. This has the effect of allowing the BRDF to be identified by color. Once the BRDF is obtained, it has the effect of being able to identify the presence or absence of minute defects on the surface of object S.

[0103] This embodiment uses a line sensor as the image sensor 44. The image sensor (line sensor) 44 has the characteristic of being able to acquire an image of the surface of an object S being transported in a predetermined direction at a predetermined speed with high accuracy. Therefore, by using the optical inspection system 10 according to this embodiment, it is possible to inspect the surface of an object S being transported with high accuracy and to acquire three-dimensional shape information of the surface of the object S.

[0104] Furthermore, in this embodiment, by making the longitudinal directions of the line sensor 44 and the illumination unit 22 parallel and elongated, it is possible to acquire images of a wide range of object surfaces S. For example, the longitudinal dimensions of the line sensor and illumination unit can be several hundred mm to several thousand mm.

[0105] In this embodiment, since the wavelength selection unit 26 is positioned in front of the imaging unit 24, this optical system can be assembled for any imaging unit (i.e., camera) 24. In other words, it has the advantage of a wide range of choices for the imaging unit (camera) 24.

[0106] Furthermore, the wavelength selection unit 26 is supported by a support unit 72, and this support unit 72 allows the wavelength selection unit 26 to be rotated by a first adjustment unit 74. In this case, there is an advantage that a highly accurate BRDF distribution can be obtained by imaging the surface of the object S in accordance with the rotation of the wavelength selection unit 26 in accordance with the rotation of the support unit 72.

[0107] According to this embodiment, an optical inspection program, an optical inspection system 10, a processing device 14 for the optical inspection system 10, and an optical inspection method can be provided that remove the image (wavelength selection pattern) of the wavelength selection unit 26 included in the image when acquiring BRDF.

[0108] In the first to fourth embodiments described above, examples were given in which each wavelength selection region 52, 54(,56) extends in the y-axis direction. For example, it is also preferable that each wavelength selection region 52, 54(,56) of the wavelength selection unit 26 is formed in an annular shape. In this case, by performing the processing described in the first embodiment on the obtained image, or by performing the Fourier transform processing, appropriate masking, and inverse Fourier transform processing described in the second and third embodiments, the color applied by the wavelength selection unit 26 can be removed while leaving scratches, etc.

[0109] According to at least one embodiment described above, it is possible to provide an optical inspection program, an optical inspection system 10, a processing device 14 for the optical inspection system 10, and an optical inspection method for removing the image (wavelength selection pattern) of the wavelength selection unit 26 included in the image when acquiring BRDF.

[0110] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. The claims of this application as they were at the time of filing are included below. [Note 1] The process involves taking an image of an object surface captured through a wavelength selector that converts incident light into at least two different wavelength spectra, and then removing the image of the wavelength selector included in the captured image to generate a wavelength selector-removed image. An optical inspection program that causes the processor to execute. [Note 2] Removing the image of the wavelength selection portion included in the captured image is, The Fourier transform is used to generate the captured image in the frequency domain. In the captured image in the frequency space, the value of the frequency component corresponding to the wavelength selection portion is kept constant and small to generate an image with the wavelength selection portion removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. The optical inspection program described in Appendix 1, including the optical inspection program described in Appendix 1. [Note 3] The optical inspection program described in Appendix 1, wherein removing the wavelength selection portion included in the captured image includes applying a spatial high-pass filter to the captured image. [Note 4] The wavelength selection unit has a periodic pattern, as described in Appendix 1 of the optical inspection program. [Appendix 5] Removing the wavelength selection portion included in the captured image means that The Fourier transform is used to generate the captured image in the frequency domain. In the frequency space, the values ​​of the frequency components corresponding to the wavelength selection area are kept constant and small at two points or in the vicinity thereof of the captured image, thereby generating an image with the wavelength selection area removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. The optical inspection program described in Appendix 4, including the optical inspection program described in Appendix 4. [Note 6] To obtain information about the object surface based on the wavelength-selective removal image, An optical inspection program as described in Appendix 1, which causes the processor to execute the above. [Note 7] A storage unit for storing the optical inspection program described in any one of Appendix 1 to Appendix 6, A processor that reads and executes the program stored in the storage unit from the storage unit. A processing device for an optical inspection system, having the following features. [Note 8] A processing device for an optical inspection system, comprising a processor that removes the image of the wavelength selector from an image of an object surface captured through a wavelength selector that converts incident light into at least two different wavelength spectra, thereby generating a wavelength selector-removed image. [Note 9] To obtain information about the object surface based on the wavelength-selective removal image, A processing apparatus for the optical inspection system described in Appendix 8, which causes the processor to perform the above. [Appendix 10] The processing apparatus described in Appendix 8 or Appendix 9, The imaging unit that acquires the aforementioned captured image and An optical inspection system equipped with the following features. [Note 11] In a cross-section including the optical axis of the imaging unit, an illumination unit is provided that emits parallel light. The optical inspection system described in Appendix 10. [Note 12] The process involves taking an image of an object surface captured through a wavelength selector that converts incident light into at least two different wavelength spectra, and then removing the image of the wavelength selector included in the captured image to generate a wavelength selector-removed image. Optical inspection methods, including those mentioned above. [Note 13] To obtain information about the object surface based on the wavelength-selective removal image, The optical inspection method described in Appendix 12, wherein the processor is made to execute the above. [Note 14] Removing the image of the wavelength selection portion included in the captured image is, The Fourier transform is used to generate the captured image in the frequency domain. In the captured image in the frequency space, the value of the frequency component corresponding to the wavelength selection portion is kept constant and small to generate an image with the wavelength selection portion removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. The optical inspection method described in Appendix 13, including the optical inspection method described in Appendix 13. [Explanation of Symbols]

[0111] 10…Optical inspection system, 12…Optical device, 14…Processing device, 22…Illumination unit, 24…Imaging unit, 26…Wavelength selection unit, 28…Beam splitter, 42…Imaging optical element, 44…Image sensor, 52…First wavelength selection region, 54…Second wavelength selection region, 56…Third wavelength selection region, 61…Processor, 72…Support unit, 74…First adjustment unit.

Claims

1. Light incident from the surface of an object is captured by a wavelength selection unit having at least two wavelength selection regions with a pattern, and the wavelength selection unit captures images of the object surface by converting the light into at least two different wavelength spectra. The image of the pattern of the wavelength selection unit is then projected onto the captured image of the object surface, and the image of the pattern of the wavelength selection unit included in the captured image is removed to generate an image of the object surface with the wavelength selection unit removed. An optical inspection program that causes the processor to execute.

2. The image of the object surface is a color image. The optical inspection program according to claim 1.

3. Removing the image of the wavelength selection portion included in the captured image is, The Fourier transform is used to generate the captured image in the frequency domain. In the captured image in the frequency space, the value of the frequency component corresponding to the wavelength selection portion is kept constant and small to generate an image with the wavelength selection portion removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. The optical inspection program according to claim 1, including the following:

4. The optical inspection program according to any one of claims 1 to 3, wherein the pattern of the wavelength selection unit is periodic.

5. Removing the wavelength selection portion included in the captured image means that The Fourier transform is used to generate the captured image in the frequency domain. In the frequency space, the values ​​of the frequency components corresponding to the wavelength selection area are kept constant and small at two points or in the vicinity thereof in the captured image, thereby generating an image with the wavelength selection area removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. The optical inspection program according to claim 4, including the following:

6. To obtain information about the object surface based on the wavelength-selective removal image, An optical inspection program according to any one of claims 1 to 3, which causes the processor to execute the following.

7. A storage unit for storing the optical inspection program described in any one of claims 1 to 3, The processor reads the optical inspection program stored in the storage unit from the storage unit and executes it. A processing device for an optical inspection system, having the following features.

8. A processing device for an optical inspection system, comprising a processor that captures light incident from the surface of an object into at least two different wavelength spectra in a wavelength selection unit having at least two or more wavelength selection regions having a pattern, and then removes the image of the pattern of the wavelength selection unit from the captured image of the object surface onto which the image of the pattern of the wavelength selection unit is projected, thereby generating an image of the object surface which is a wavelength selection unit removed image.

9. The image of the object surface is a color image. The apparatus according to claim 8.

10. To obtain information about the object surface based on the wavelength-selective removal image, The processing apparatus according to claim 8, wherein the processor is made to execute the above.

11. The processing apparatus according to any one of claims 8 to 10, An imaging unit that acquires the image of the object surface onto which the image of the pattern of the wavelength selection unit is projected. An optical inspection system equipped with the following features.

12. In a cross-section including the optical axis of the imaging unit, an illumination unit is provided that emits parallel light. The optical inspection system according to claim 11.

13. Light incident from the surface of an object is captured by a wavelength selection unit having at least two wavelength selection regions with a pattern, and the wavelength selection unit captures images of the object surface by converting the light into at least two different wavelength spectra. The image of the pattern of the wavelength selection unit is then projected onto the captured image of the object surface, and the image of the pattern of the wavelength selection unit included in the captured image is removed to generate an image of the object surface with the wavelength selection unit removed. Optical inspection methods, including those mentioned above.

14. The image of the object surface is a color image. The optical inspection method according to claim 13.

15. To obtain information about the object surface based on the wavelength-selective removal image, The optical inspection method according to claim 13, including the method described in claim 13.

16. Removing the image of the wavelength selection portion included in the captured image is, The Fourier transform is used to generate the captured image in the frequency domain. In the captured image in the frequency space, the value of the frequency component corresponding to the wavelength selection portion is kept constant and small to generate an image with the wavelength selection portion removed in the frequency space. Performing an inverse Fourier transform on the wavelength-selective-removed image in the frequency space generates the wavelength-selective-removed image in real space. An optical inspection method according to any one of claims 13 to 15, including the method described in any one of claims 13 to 15.

Images