Optical inspection apparatus, optical inspection method, and optical inspection program

Abstract

To provide an optical inspection apparatus, which is capable of acquiring information regarding an object with illumination of a small number of colors (wavelength spectrum).SOLUTION: An optical inspection apparatus 10 includes an illumination section 20, an imaging section 30, and a processing section 40. The illumination section is configured to irradiate at least an object point of an object with a light beam flux at one or a plurality of solid angles. The processing section causes an object point to be irradiated with the light beam flux at a first solid angle A1 by first illumination light from the illumination section, sets a solid angle not including the first solid angle as a second solid angle A2, and causes a first captured image of an object to be acquired using the imaging section on the basis of illumination by the first illumination light. The processing section sets a solid angle included in the first solid angle as a third solid angle A3, sets a solid angle included in the second solid angle as a fourth solid angle A4, causes an object point to be irradiated with a light beam flux of the third solid angle and the fourth solid angle by second illumination light from the illumination section, causes a second captured image of the object to be acquired using the imaging section on the basis of illumination by the second illumination light, and acquires information of the object on the basis of the first captured image and the second captured image.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] An embodiment of the present invention relates to an optical inspection device, an optical inspection method, and an optical inspection program. [Background technology]

[0002] Non-contact inspection of objects is becoming increasingly important in various industries. Conventional methods involve using a diffraction grating or wavelength filter to separate light beams, creating a one-to-one correspondence between the color (wavelength spectrum) and the direction of the beam, and identifying the color to identify the direction of the beam, thereby obtaining information about the surface or interior of the object. [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] H. Ohno, “One-shot BRDF imaging system to obtain surface properties,” Volume 28, pages 655-661, (2021) [Non-patent document 3] H. Ohno, “One-shot color mapping of a ray direction field for obtaining three-dimensional profiles integrating deep neural networks,” Vol. 2, No. 9, Optics Continuum (2023). Summary of the Invention [Problem to be solved by the invention]

[0005] The problem to be solved by the present invention is to provide an optical inspection device, an optical inspection method, and an optical inspection program that can acquire information about an object using illumination of a limited number of colors (wavelength spectrum). [Means for solving the problem]

[0006] According to an embodiment, an optical inspection device includes an illumination unit, an imaging unit, and a processing unit. The illumination unit irradiates at least a first object point of the object with ray bundles of one or more solid angles, respectively. The imaging unit acquires an image of the object in response to illumination with the ray bundles of one or more solid angles. The processing unit irradiates the first object point of the object with ray bundles of a first solid angle formed by a first illumination light from the illumination unit, defines a solid angle not including the first solid angle as a second solid angle, and causes the imaging unit to acquire a first captured image of the object based on illumination with the first illumination light. The processing unit also irradiates at least the first object point with ray bundles of the third and fourth solid angles formed by a second illumination light from the illumination unit, and causes the imaging unit to acquire a second captured image of the object based on illumination with the second illumination light. Then, the processing unit acquires information about the object based on the first captured image and the second captured image. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a schematic diagram showing an optical inspection device according to a first embodiment. [Figure 2] FIG. 1 is a schematic block diagram of an optical inspection apparatus according to an embodiment. [Figure 3] 4 is a processing flow when acquiring information about an object using the optical inspection device according to the first embodiment. [Figure 4] FIG. 10 is a schematic diagram showing an optical inspection device according to a first modified example of the first embodiment. [Figure 5]10 is a processing flow when acquiring information about an object using an optical inspection device according to Modification 2 of the first embodiment. [Figure 6] FIG. 10 is a schematic diagram showing an optical inspection device according to a second embodiment. [Figure 7] FIG. 10 is a schematic diagram showing an optical inspection device according to a third embodiment. [Figure 8] FIG. 10 is a schematic diagram showing an optical inspection device according to a fourth embodiment. [Figure 9] 13A and 13B are schematic diagrams showing projection patterns from an illumination unit of an optical inspection device according to a fourth modification of the fourth embodiment, in which (A) shows a first projection pattern and (B) shows a second projection pattern. [Figure 10] 13A and 13B are schematic diagrams showing projection patterns from an illumination unit of an optical inspection device according to a fifth modified example of the fourth embodiment, in which (A) shows a first projection pattern and (B) shows a second projection pattern. [Figure 11] 13A and 13B are schematic diagrams showing projection patterns from an illumination unit of an optical inspection device according to a sixth modified example of the fourth embodiment, in which (A) shows a first projection pattern and (B) shows a second projection pattern. [Figure 12] 13A and 13B are schematic diagrams showing projection patterns from an illumination unit of an optical inspection device according to a seventh modified example of the fourth embodiment, in which (A) shows a first projection pattern and (B) shows a second projection pattern. [Figure 13] FIG. 10 is a schematic diagram showing an application example 1 in which an optical inspection of an object is performed using a projection pattern according to Modification 4 of the fourth embodiment in the optical inspection device according to the fourth embodiment. [Figure 14] FIG. 10 is a schematic diagram showing a second application example when an optical inspection of an object is performed using the projection pattern of the fourth modification of the fourth embodiment in the optical inspection device according to the fourth embodiment. [Figure 15] FIG. 10 is a schematic diagram showing an application example 3 in which an optical inspection of an object is performed using a projection pattern according to a fourth modification of the fourth embodiment in the optical inspection device 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, etc., are not necessarily the same as those in reality. Furthermore, even when the same part is shown, the dimensions and ratios may be different depending on the drawing. In this specification and each drawing, elements similar to those previously described with reference to the previous drawings are designated by the same reference numerals, and detailed explanations will be omitted where appropriate.

[0009] In this specification, light is a type of electromagnetic wave, and includes gamma rays, X-rays, ultraviolet rays, visible light, infrared rays, radio waves, etc. In this embodiment, light is considered to be visible light, with a wavelength in the range of 400 nm to 800 nm, for example.

[0010] (First embodiment) An optical inspection device 10 according to this embodiment will be described below with reference to FIGS.

[0011] Fig. 1 shows a schematic cross-sectional view of an optical inspection device 10 according to this embodiment, and Fig. 2 shows a schematic block diagram of the optical inspection device 10 according to this embodiment.

[0012] As shown in FIGS. 1 and 2, the optical inspection device 10 according to this embodiment includes an illumination unit 20, an imaging unit 30, and a processing unit 40.

[0013] The illumination unit 20 can emit light of at least a first wavelength spectrum from a light source. A wavelength spectrum refers to the intensity distribution of light relative to wavelength. Two different wavelength spectra mean that the intensity distributions relative to wavelength are different from each other. For example, a wavelength spectrum having a peak at a wavelength of 450 nm is different from a wavelength spectrum having a peak at a wavelength of 650 nm. The first wavelength spectrum emitted by the illumination unit 20 may be any type. In this embodiment, the first wavelength spectrum is white light having a peak wavelength of 550 nm.

[0014] The light source of the illumination unit 20 may be a white light source such as a white LED (Light-Emitting Diode), a halogen lamp, a fluorescent lamp, an incandescent lamp, a high-intensity discharge (HID) lamp, or a metal halide lamp. In this embodiment, the light source is a white LED. However, the light source is not limited to this and may also be a monochromatic laser. Alternatively, it may be an array of multiple monochromatic lasers of various colors.

[0015] The imaging unit 30 uses an image sensor 32 and an imaging optical element 34, and can image light onto the image sensor 32 using the imaging optical element 34. The imaging optical element 34 may be, for example, a single lens, a compound lens made up of multiple lenses, a Fresnel lens, a fly's eye lens, a microlens array, a concave mirror, a diffraction grating, a gradient index lens (GRIN lens), or the like. In other words, anything that can image light is acceptable. The plane on which a set of points at infinity are imaged by the imaging optical element 34 is called a focal plane f 30 The optical axis C of the imaging optical element 34 30 is the focal plane f 30 The imaging optical element 34 is a straight line perpendicular to the line 31, and light emitted from a point on the straight line is imaged again on the straight line. In this embodiment, the imaging optical element 34 is a compound lens.

[0016] In this specification, the imaging optical element 34 for imaging is specifically referred to as the imaging optical element, and the imaging optical element 24 for illumination is referred to as the illumination imaging optical element. 30 and the optical axis of the illumination imaging optical element 24 is the illumination optical axis C 20 Furthermore, let each focal plane be the imaging focal plane f 30 and illumination focal plane f 20 Let's say.

[0017] The object P may be either light-transmitting or light-reflecting. Alternatively, the object P may be translucent. A point on the surface of the object P or inside the object P is called an object point. In the following, unless otherwise specified, the object P is assumed to reflect light, and the object point is assumed to be on the surface of the object P. The surface of the object P may also be called the object surface or object surface. In this embodiment, the object P is assumed to be reflective, and light is reflected by the surface. Therefore, the object point is present on the surface of the object. However, this is not limited to the object P.

[0018] In this embodiment, white light is used as the illumination light as described above. The image sensor 32 according to this embodiment may be, for example, a monochrome camera that can detect black and white as a difference in light intensity, or a color camera.

[0019] The processing unit 40 controls the illumination unit 20 and the imaging unit 30. The processing unit 40 is, for example, a computer. The processing unit 40 includes a processor or integrated circuit (control circuit) including a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), and a storage medium such as a memory. The processing unit 40 may include one or more processors or integrated circuits. The processing unit 40 performs processing by executing a program or the like stored in a storage medium or the like. For example, one example of the program is an optical inspection program for the object P.

[0020] In addition, in the processing unit 40, the program executed by the processor may be stored in a computer (server) connected via a network such as the Internet or a server in a cloud environment, etc. In this case, the processing unit 40 downloads the program via the network.

[0021] The processing unit 40 may be located at a location remote from the optical inspection device 10, whether domestic or overseas, as long as it can appropriately control the illumination unit 20 and the imaging unit 30.

[0022] With the above-described configuration, the operation of the optical inspection device 10 according to this embodiment will be described using the processing flow shown in FIG.

[0023] The processing unit 40 causes the illumination unit 20 to irradiate light of a first wavelength spectrum as first illumination light onto at least a first object point P1 on the object plane P. The processing unit 40 also causes the imaging unit 30 to capture an image at a viewpoint sufficiently far away from the object P. That is, the processing unit 40 causes the imaging unit 30 to capture an image of the first illumination light from the object P. At this time, the light beam captured by the imaging unit 30 is aligned along the imaging optical axis C. 30 can be considered to be nearly parallel.

[0024] However, the processing unit 40 is not limited to this, and may instead cause the illumination unit 20 to irradiate light of a first wavelength spectrum as first illumination light onto at least a first object point P1 on the object plane P. Then, immediately thereafter, the illumination unit 20 may cause the imaging unit 30 to capture an image from a viewpoint sufficiently far away from the object P. In this case, the processing unit 40 controls only the illumination unit 20. Even in this case, the processing unit 40 causes the first illumination light from the object P to be captured. This type of control has the advantage of enabling a series of sequential operations, i.e., illumination and imaging, to be performed reliably and quickly.

[0025] That is, the processing unit 40 first causes the illumination unit 20 to irradiate a first object point P1 on the object plane P with light of a first wavelength spectrum as first illumination light so that a bundle of light (luminous flux) forms a first solid angle A1 (step S11). A solid angle not including the first solid angle A1 is defined as a second solid angle A2 (all solid angles other than the first solid angle A1). Then, the processing unit 40 causes the imaging unit 30 to capture an image of the object P using the first illumination light, thereby acquiring a first captured image (step S21).

[0026] The processing unit 40 causes the illumination unit 20 to irradiate light of a first wavelength spectrum as second illumination light onto at least a first object point P1 on the object plane P. The processing unit 40 also causes the imaging unit 30 to acquire an image from a viewpoint sufficiently far away from the object P. In other words, the processing unit 40 causes the object P to be imaged using the second illumination light.

[0027] That is, the processing unit 40 causes the illumination unit 20 to irradiate the first object point P1 with light of the first wavelength spectrum as the second illumination light so that the light beam forms a third solid angle A3 and a fourth solid angle A4 at the first object point P1 (step S11). The third solid angle A3 is included in the first solid angle A1. At the same time, the fourth solid angle A4 is included in the second solid angle A2. Here, the third solid angle A3 and the fourth solid angle A4 are assumed to be continuous regions. However, they do not have to be continuous and may be discontinuous. Then, the processing unit 40 causes the imaging unit 30 to image the object P using the second illumination light and acquire a second captured image (step S21). Note that, for ease of explanation, in the cross section of FIG. 1, the cross-sectional area of ​​the region defined by the third solid angle A3 is assumed to be half the cross-sectional area of ​​the region defined by the first solid angle A1, and the region defined by the second solid angle A2 is assumed to include the region defined by the fourth solid angle A4. The second solid angle A2 may be the entire solid angle not included in the first solid angle A1, or a smaller solid angle. In other words, the second solid angle A2 may be any solid angle not included in the first solid angle A1. However, this is not limited to this, and the second solid angle A2 and the fourth solid angle A4 may be the same, that is, they may be common. Assume that the cross-sectional areas of the area defined by the third solid angle A3 and the area defined by the fourth solid angle A4 are the same.

[0028] The order of acquiring the first captured image and the second captured image may be reversed. Therefore, the processing unit 40 can irradiate at least a first object point P1 of the object P with a bundle of rays of the first wavelength spectrum at a first solid angle A1 as the first illumination light of the illumination unit 20, and before or after that, can irradiate at least the first object point P1 with bundles of rays of the first wavelength spectrum at a third solid angle A3 and a fourth solid angle A4 as the second illumination light of the illumination unit 20. Therefore, the processing unit 40 can irradiate the first illumination light and the second illumination light at different times and acquire captured images for each.

[0029] For example, consider a case where the surface of the object P is smooth. In this case, light incident on the surface of the object P is specularly reflected. Here, the light reflected from the object point P on the surface of the object P is reflected along the imaging optical axis C. 30 and is captured by the imaging unit 30. In other words, the direction of the light ray just before being captured by the imaging unit 30 is the imaging optical axis C 30 . In this case, there is a one-to-one relationship between the angle of incidence of light incident on object point P1 and the tilt angle of the surface of object P at object point P1 on the incident plane (cross section including the incident light). In other words, the tilt angle can be determined by determining the angle of incidence, and vice versa. Therefore, the tilt angle converted from the angle of incidence is called the converted tilt angle. The method for calculating the converted tilt angle may vary depending on the spatial arrangement of the illumination unit 20 and the image capture unit 30, but the one-to-one relationship between the angle of incidence and the tilt angle remains the same. The calculation of the converted tilt angle is determined by the light source and optical element configurations used in the illumination unit 20, the optical element configurations used in the image capture unit 30, and the distances from the illumination unit 20 and the image capture unit 30 to the object P. However, even if the surface of object P is not smooth, the angular distribution of reflected light generally exhibits a stronger specular reflection component (intensity distribution in the specular reflection direction and its vicinity). The one-to-one relationship described above holds for the specular reflection component.

[0030] Suppose a first object point P1 is captured in the first captured image. In this case, the tilt angle θ can be identified as being within the region (range) of converted tilt angles converted from the first solid angle A1, and this region is called the first converted tilt angle region. On the other hand, suppose the first object point P1 is not captured in the first captured image. In this case, it is found that the tilt angle θ is within a region (range) excluding the first converted tilt angle. This is called the second converted tilt angle region.

[0031] Assume that a first object point P1 appears in the first captured image and also appears in the second captured image. In this case, the tilt angle θ can be identified as a region of converted tilt angles converted from the third solid angle A3, and this region is designated as the third converted tilt angle region. Because the third solid angle A3 is included in the first solid angle A1, the third converted tilt angle region is narrower than the first converted tilt angle region. The angle at the first object point P1 in the third converted tilt angle region is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the tilt of the surface of the object P on which the first object point P1 exists. That is, the processing unit 40 acquires information about the object P at the first object point P1 (step S3).

[0032] Also, suppose that the first object point P1 appears in the first captured image, but does not appear in the second captured image. In this case, the tilt angle can be determined as the area obtained by excluding the third tilt angle area from the first converted tilt angle area. This is called the third converted tilt angle complementary area. Because the third solid angle A3 is included in the first solid angle A1, this third converted tilt angle complementary area is narrower than the first converted tilt angle area. The angle at the first object point P1 in the third converted tilt angle complementary area is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the first object point P1 (step S3).

[0033] On the other hand, suppose that the first object point P1 is not captured in the first captured image, but is captured in the second captured image. In this case, the tilt angle θ can be identified as a region of converted tilt angles converted from the fourth solid angle A4, and this region is designated as the fourth converted tilt angle region. Because the fourth solid angle A4 is included in the second solid angle A2, this fourth converted tilt angle region is narrower than the second converted tilt angle region. The angle at the first object point P1 in the fourth converted tilt angle region is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the first object point P1 (step S3).

[0034] Also, suppose that the first object point P1 is not captured in the first captured image and is not captured in the second captured image. In this case, the tilt angle can be determined as the area obtained by excluding the fourth tilt angle area from the second converted tilt angle area. This area is called the fourth converted tilt angle complementary area. Because the fourth solid angle A4 is included in the second solid angle A2, this fourth converted tilt angle complementary area is narrower than the second converted tilt angle area. The angle at the first object point P1 in the fourth converted tilt angle complementary area is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the first object point P1 (step S3).

[0035] As a result, by using the first captured image and the second captured image in combination, the tilt angle θ at the first object point P1 becomes an estimated tilt angle in one of the four tilt angle regions: the third converted tilt angle region, the third converted tilt angle complementary region, the fourth converted tilt angle region, and the fourth converted tilt angle complementary region. Therefore, the processing unit 40 acquires information about the first object point P1 of the object P (step S3).

[0036] In the above example, the estimated tilt angle is calculated in step S3. In step S3, the converted tilt angle θ may be output in each loop other than the final loop. In this case, the estimated tilt angle may be calculated in step S3 of the final loop, for example.

[0037] The tilt angle θ obtained from only the first captured image can fall within either the first converted tilt angle region or the second converted tilt angle region, depending on whether the first object point P1 is captured. Here, the first converted tilt angle region is larger than the third converted tilt angle region or the third converted tilt angle complementary region. The second converted tilt angle region is larger than the fourth converted tilt angle region or the fourth converted tilt angle complementary region. This means that the tilt angle θ (information about object P) obtained by combining the first and second captured images can be estimated with higher accuracy than the tilt angle θ (information about object P) obtained from only the first captured image.

[0038] On the other hand, the tilt angle θ obtained from only the second captured image can be a region that combines the third converted tilt angle region and the fourth converted tilt angle region, or a region that does not include them, depending on whether the first object point P1 is captured. The region that combines the third converted tilt angle region and the fourth converted tilt angle region is larger than the third converted tilt angle region or the fourth converted tilt angle region. Also, the region that does not include the region that combines the third converted tilt angle region and the fourth converted tilt angle region is larger than the third converted tilt angle region or the fourth converted tilt angle region. This means that the tilt angle θ (information about object P) obtained by combining the first captured image and the second captured image can be estimated with higher accuracy than the tilt angle θ obtained from only the second captured image.

[0039] In this way, the optical inspection device 10 according to this embodiment has the advantage of being able to obtain more detailed information about the tilt angle θ of the object P (information about the object P) by combining the first captured image and the second captured image. In other words, by combining the first captured image and the second captured image as in the method of the optical inspection device 10 according to this embodiment shown in Fig. 3, the range (area) in which the tilt angle θ can be obtained can be narrowed compared to the range obtained from only one of the captured images.

[0040] Note that i and n in FIG. 3 are natural numbers, and it is preferable that n is 2 or greater and i is a natural number less than or equal to n. In this way, by repeatedly using the method of the optical inspection device 10 according to this embodiment inductively, it is possible to improve the estimation accuracy of the tilt angle θ. For example, after acquiring the second captured image, a fifth solid angle and a sixth solid angle are formed with respect to the third solid angle A3 so that the third solid angle A3 and the fourth solid angle A4 have the same relationship with respect to the first solid angle A1. Then, a seventh solid angle and an eighth solid angle are formed with respect to the fourth solid angle A4 so that the third solid angle A3 and the fourth solid angle A4 have the same relationship with respect to the first solid angle A1. The fifth solid angle, the sixth diagonal angle, the seventh solid angle, and the eighth solid angle are irradiated with third illumination light (light of the first wavelength spectrum). A third captured image is then acquired using the imaging unit 30 based on illumination with the third illumination light.

[0041] Furthermore, for the fifth and sixth solid angles, the ninth and tenth solid angles, and the eleventh and twelfth solid angles are formed, respectively, in the same manner as the relationship between the third solid angle A3 and the fourth solid angle A4 relative to the first solid angle A1. Furthermore, for the eighth and ninth solid angles, the thirteenth and fourteenth solid angles, and the fifteenth and sixteenth solid angles are formed, respectively, in the same manner as the relationship between the third solid angle A3 and the fourth solid angle A4 relative to the first solid angle A1. The ninth to sixteenth solid angles are irradiated with fourth illumination light (light of the first wavelength spectrum), and a fourth captured image is acquired using the imaging unit 30 based on the illumination with the fourth illumination light.

[0042] The optical inspection device 10 according to this embodiment can capture more images by dividing the solid angle into smaller increments, and can obtain information about the object based on these captured images. In this way, the optical inspection device 10 according to this embodiment exponentially increases the number of solid angles within a predetermined range (the range of the first solid angle A1 and the second solid angle A2 in FIG. 1 ). The processing unit 40 acquires captured images accordingly, determines whether the object point P1 is captured, and improves the accuracy of estimating the tilt angle θ accordingly. The optical inspection device 10 according to this embodiment uses only the first wavelength spectrum, i.e., white light. Even when using only one color, the optical inspection device 10 according to this embodiment can be used recursively (or iteratively) to improve the accuracy of obtaining information about the object P.

[0043] For example, in the cross section shown in FIG. 1 , the cross-sectional area of ​​a sector formed by the third solid angle A3 and centered on the first object point P1 by the first solid angle A1 is half that of the cross-sectional area of ​​a sector formed by the third solid angle A3 and centered on the first object point P1. Similarly, when the cross-sectional areas of the sectors formed by the third solid angle A3 and the fourth solid angle A4 (second solid angle A2) are the same, the accuracy of the tilt angle θ that the processing unit 40 can estimate from the first captured image and the second captured image is, for example, twice as high as when only the first captured image is acquired. Here, "double the accuracy" means that the range of possible estimated tilt angles is halved. Furthermore, as described above, in this embodiment, the accuracy of the tilt angle θ that can be estimated by further using the third captured image obtained by forming the fifth, sixth, seventh, and eighth solid angles is four times as high as when only the first captured image is acquired. Furthermore, the accuracy of the tilt angle θ that can be estimated using a fourth captured image obtained by forming a solid angle in a similar manner is 16 times higher than when only the first captured image is obtained. In this way, the optical inspection device 10 according to this embodiment can more accurately determine the tilt angle θ of the target object point by making the corresponding solid angle finer and obtaining the captured images corresponding to them.

[0044] The range of the first wavelength spectrum, which is the illumination light used in this embodiment, may be appropriately wide, such as white light, or may be narrower than that. An example of a narrow wavelength range of the first wavelength spectrum is a single color such as blue, red, or green, and can be set appropriately.

[0045] In this embodiment, after forming a bundle of rays at the first solid angle A1, the illumination unit 20 simultaneously forms bundles of rays at the third solid angle A3 and the fourth solid angle A4. Meanwhile, consider a case where only a bundle of rays at either the third solid angle A3 or the fourth solid angle A4 is formed.

[0046] For example, consider a case where only a bundle of rays forming the third solid angle A3 is formed. In this case, when the first captured image and the second captured image are combined and used, the processing unit 40 determines that the tilt angle θ at the first object point P1 falls into one of three tilt angle regions: the second converted tilt angle region, the third converted tilt angle region, and the third converted tilt angle complementary region. Here, the second converted tilt angle region is larger than the fourth converted tilt angle region or the fourth converted tilt angle complementary region. In other words, in this case, the range of possible tilt angles θ is wider than when the optical inspection device 10 of this embodiment is used. In other words, the estimation accuracy of the tilt angle θ is lower than when the optical inspection device 10 of this embodiment is used.

[0047] For example, consider a case where only a bundle of rays forming a fourth solid angle A4 is formed. In this case, when the first captured image and the second captured image are combined and used, the processing unit 40 determines that the tilt angle θ at the first object point P1 falls into one of three tilt angle regions: the first converted tilt angle region, the fourth converted tilt angle region, and the fourth converted tilt angle complementary region. Here, the first converted tilt angle region is larger than either the third converted tilt angle region or the third converted tilt angle complementary region. In other words, in this case, the range of possible tilt angles θ is wider than when the optical inspection device 10 of this embodiment is used. In other words, the estimation accuracy of the tilt angle θ is lower than when the optical inspection device 10 of this embodiment is used.

[0048] As described above, in the optical inspection device 10 of this embodiment, the solid angle included in the first solid angle A1 is defined as the third solid angle A3, the solid angle included in the second solid angle A2 is defined as the fourth solid angle A4, and the illumination unit 20 simultaneously forms ray bundles of the third solid angle A3 and the fourth solid angle A4, thereby achieving the effect of improving the estimation accuracy of the tilt angle θ.

[0049] The directional distribution of reflected light from the surface of object P changes depending on the surface texture and shape. The directional distribution of reflected light is described by BRDF (Bidirectional Reflectance Distribution Function). Generally, the BRDF can be used to estimate the surface texture and shape of the surface of object P, that is, information about the object surface P. When the surface of object P is smooth, light incident on that surface is specularly reflected. In this case, the BRDF indicates that there is only one specular reflection direction for one incident direction. This means that, for example, if the specular reflection direction is along the imaging optical axis C, 30 If it is determined that the incident angle is parallel to the surface of the object P, then it can be said that there is a one-to-one relationship between the inclination angle θ of the surface of the object P and the angle of incidence. On the other hand, if the surface of the object P is rough, the light that is incident on that surface is scattered. At this time, a diffuse component is also generated in addition to the specular component. However, the specular component generally has a greater light intensity than the diffuse component.

[0050] In the above, we considered an example in which the object surface P is smooth and specularly reflects light. However, this is not limiting, and the object surface P may also be a rough surface. In this case, the specular reflection component has a greater light intensity than the diffuse component. In the above discussion, when the first object point P1 is captured, the processing unit 40 determines whether the first object point P1 is captured in the captured image. However, here, the processing unit 40 may also determine whether the light intensity of the first object point P1 is high or low. For example, the processing unit 40 may use the difference in light intensity between white and black light received by the imaging unit 30. By performing such a distinction, the processing unit 40 can obtain a similar conclusion even when the object surface P is a rough surface. In other words, the processing unit 40 combines the first captured image and the second captured image and compares the light intensity of the captured image of the first object point P1, thereby improving the accuracy of estimating the tilt angle θ of the object surface P. Therefore, this embodiment has the effect of acquiring information about the shape of the object surface P having various BRDFs. Of course, the processing unit 40 may estimate the inclination angle θ of the object plane P as information about the object P by combining the first captured image and the second captured image, comparing the magnitude of the light intensity of the captured image of the first object point P1, and determining whether the first object point P1 is captured.

[0051] In the present embodiment, an example of acquiring information about the object P has been described using the surface of the object P as an example. For example, it is also possible to acquire information about the inner surface of the object P instead of the surface.

[0052] Therefore, the optical inspection device 10 according to this embodiment includes an illumination unit 20, an imaging unit 30, and a processing unit 40. The illumination unit 20 irradiates bundles of rays of multiple solid angles onto at least a first object point P1 of the object P. Note that multiple solid angles here include a single solid angle. The imaging unit 30 acquires an image of the object P in response to illumination by bundles of rays of one or multiple solid angles. The processing unit 40 irradiates a first object point P1 of the object P with a bundle of rays from a first solid angle A1 due to the first illumination light from the illumination unit 20, defines a solid angle not including the first solid angle A1 as a second solid angle A2, causes the imaging unit 30 to acquire a first captured image of the object P based on illumination with the first illumination light, defines a solid angle included in the first solid angle A1 as a third solid angle A3, and defines a solid angle included in the second solid angle A2 as a fourth solid angle A4, causes the processing unit 40 to irradiate at least the first object point P1 with bundles of rays from the third solid angle A3 and the fourth solid angle A4 due to the second illumination light from the illumination unit 20, causes the imaging unit 30 to acquire a second captured image of the object P based on illumination with the second illumination light, and acquires information about the object P based on the first and second captured images.

[0053] Furthermore, the optical inspection method according to this embodiment includes irradiating a bundle of rays from a first illumination light from the illumination unit 20 at a first solid angle A1 onto at least a first object point P1 of the object P, wherein a solid angle not including the first solid angle A1 is defined as a second solid angle A2; acquiring a first captured image of the object P using the imaging unit 30 based on illumination by the first illumination light; defining a solid angle included in the first solid angle A1 as a third solid angle A3, and a solid angle included in the second solid angle A2 as a fourth solid angle A4; irradiating at least the first object point P1 with bundles of rays from the third solid angle A3 and the fourth solid angle A4 of the second illumination light from the illumination unit 20; acquiring a second captured image of the object P using the imaging unit 30 based on illumination by the second illumination light; and acquiring information about the object P based on the first captured image and the second captured image.

[0054] In addition, the optical inspection program according to this embodiment causes a computer to execute the following operations: irradiating a bundle of rays from a first solid angle of a first illumination light from the illumination unit 20 onto at least a first object point of the object; defining a solid angle that does not include the first solid angle as a second solid angle; acquiring a first captured image of the object P using the imaging unit 30 based on illumination with the first illumination light; defining a solid angle included in the first solid angle A1 as a third solid angle A3 and a solid angle included in the second solid angle A2 as a fourth solid angle A4; irradiating a bundle of rays from the third solid angle A3 and the fourth solid angle A4 of the second illumination light from the illumination unit 20 onto at least the first object point P1; acquiring a second captured image of the object P using the imaging unit 30 based on illumination with the second illumination light; and acquiring information about the object P based on the first captured image and the second captured image.

[0055] Therefore, according to this embodiment, it is possible to provide an optical inspection device, an optical inspection method, and an optical inspection program that are capable of acquiring information about an object using illumination of a limited number of colors (wavelength spectrum).

[0056] In the present embodiment, for example, the first to fourth illumination lights are described as being irradiated with light of the first wavelength spectrum. For example, the wavelength spectrum of the third and fourth illumination lights may be different from that of the first and second illumination lights.

[0057] (Variation 1) In the above-described embodiment, the third solid angle A3 and the fourth solid angle A4 are continuous (connected). As shown in FIG. 4, the third solid angle A3 and the fourth solid angle A4 may be separated from each other.

[0058] In this case as well, the third solid angle A3 is included in the first solid angle A1, and the fourth solid angle A4 is included in the second solid angle A2.

[0059] Then, the processing unit 40 can combine the first captured image and the second captured image and, for example, determine whether the first object point P1 is captured, thereby estimating the inclination angle θ of the object plane P as information about the object P.

[0060] (Variation 2) In this modification, a color image sensor 32 is used as the image sensor 32. The first illumination light irradiates light of a first wavelength spectrum, and the second illumination light irradiates light of a second wavelength spectrum. The light of the first wavelength spectrum and the light of the second wavelength spectrum have different wavelength spectra that are not similar to each other.

[0061] In the example of the first embodiment described above, the processing unit 40 causes the illumination unit 20 to irradiate a bundle of rays of the first wavelength spectrum at the first solid angle A1, and then irradiates a bundle of rays of the first wavelength spectrum at the third solid angle A3 and the fourth solid angle A4.

[0062] The processing of the second modification will be briefly described with reference to the flow shown in FIG.

[0063] As shown in FIG. 5, the processing unit 40 irradiates the object P with a bundle of rays of the first wavelength spectrum at a first solid angle A1 and simultaneously irradiates the object P with bundles of rays of the second wavelength spectrum at a third solid angle A3 and a fourth solid angle A4 (step S12). Furthermore, the processing unit 40 distinguishes between the first wavelength spectrum and the second wavelength spectrum using at least two different color channels of the color image sensor 32, and acquires the respective captured images as a first captured image and a second captured image (step S22). In this manner, multiple captured images can be acquired more quickly than when irradiating illumination light of the same wavelength and sequentially acquiring the first captured image, the second captured image, and so on. Therefore, the optical inspection device 10 of this second modification has the advantage of being able to acquire information about the object P more quickly.

[0064] The solid angle included in the first solid angle A1 is defined as the third solid angle A3, and the solid angle included in the second solid angle A2 is defined as the fourth solid angle A4. The illumination unit 20 simultaneously forms a bundle of rays for the first solid angle A1 using the first wavelength spectrum and a bundle of rays for the third solid angle A3 and the fourth solid angle A4 using the second wavelength spectrum, thereby improving the accuracy of estimating the tilt angle θ. Here, the third solid angle A3 and the fourth solid angle A4 may be continuous areas. Alternatively, they may be discontinuous, independent areas.

[0065] Furthermore, similar to the process illustrated in FIG. 3 , the present method can be applied recursively and repeatedly to improve the accuracy of estimating the tilt angle θ. For example, after acquiring the first captured image and the second captured image, ray bundles of the first wavelength spectrum may be formed at the third solid angle A3, similar to the third solid angle A3 and fourth solid angle A4 relative to the first solid angle A1. Furthermore, ray bundles of the first wavelength spectrum may be formed at the seventh and eighth solid angles relative to the fourth solid angle A4, similar to the third solid angle A3 and fourth solid angle A4 relative to the first solid angle A1. Then, the fifth, sixth, seventh, and eighth solid angles may be illuminated with first illumination light having the first wavelength spectrum, and a third captured image may be acquired using the imaging unit 30 based on the illumination by the first illumination light. By doing so, the number of solid angles can be exponentially increased, and the estimation accuracy of the tilt angle θ can be improved accordingly. In this embodiment, only the first wavelength spectrum and the second wavelength spectrum, i.e., two colors, are used. Even when only two colors are used in this way, by using this embodiment inductively (or iteratively), it is possible to obtain the effect of improving the accuracy of obtaining information about the object P.

[0066] Although not explained further, if the imaging unit 30 can acquire the images as separate images, the processing unit 40 may estimate information about the object P (tilt angle θ) using illumination light of three or more different wavelength spectra.

[0067] In this modification, for example, light of a first wavelength spectrum is irradiated as the first illumination light, and light of a second wavelength spectrum is irradiated as the second illumination light. For example, the wavelength spectra of the third illumination light and the fourth illumination light may be the same as those of the first illumination light and the second illumination light, respectively, or may be different from those of both the light of the first wavelength spectrum and the light of the second wavelength spectrum.

[0068] (Second embodiment) The optical inspection device according to this embodiment will be described below with reference to Fig. 6. This embodiment is a further modification of the first embodiment including various modifications, and the same components as those described in the first embodiment or components having the same functions are assigned the same reference numerals as much as possible, and detailed descriptions thereof will be omitted.

[0069] FIG. 6 shows a schematic cross-sectional view of the optical inspection device 10 according to this embodiment.

[0070] The operation of the optical inspection device 10 according to this embodiment will be described.

[0071] The processing unit 40 causes the illumination unit 20 to simultaneously irradiate at least a first object point P1 and a second object point P2 on the object plane P with light of the first wavelength spectrum (step S12). The processing unit 40 also causes the imaging unit 30 to acquire an image at a viewpoint sufficiently far away from the object P (step S22). At this time, the light rays acquired by the imaging unit 30 are aligned along the imaging optical axis C. 30 can be considered to be nearly parallel.

[0072] In step S12 of the processing flow shown in FIG. 5, the illumination unit 20 irradiates a first object point P1 and a second object point P2 on the object plane P with light of a first wavelength spectrum as the first illumination light so that the bundle of light (luminous flux) forms a first solid angle A1. A solid angle that does not include the first solid angle A1 is defined as a second solid angle A2. The illumination unit 20 also irradiates the first object point P1 and the second object point P2 with light of a second wavelength spectrum as the second illumination light so that the bundle of light forms a third solid angle A3 and a fourth solid angle A4. The third solid angle A3 is included in the first solid angle A1. At the same time, the fourth solid angle A4 is included in the second solid angle A2.

[0073] For example, consider a case where the surface of the object P is smooth. In this case, light incident on the object surface P is specularly reflected. Here, the light reflected from an object point on the object surface P is reflected along the imaging optical axis C. 30 Assume that the light travels along the line θ and is imaged by the imaging unit 30 (step S22). Then, there is a one-to-one relationship between the angle of incidence of the light incident on the object point and the tilt angle θ of the object surface P at that object point. In other words, if the angle of incidence is determined, the tilt angle θ can be determined, and vice versa. Therefore, the tilt angle θ converted from the angle of incidence is called the converted tilt angle. However, even if the surface of the object P is not smooth, the angular distribution of the reflected light generally tends to have a strong specular reflection component. The one-to-one relationship described above holds for that specular reflection component.

[0074] The first object point P1 is the same as in the first embodiment, and by combining the first captured image and the second captured image, it is possible to obtain information about the object P at the first object point P1 (step S3).

[0075] In the following, the second object point P2 will be described.

[0076] In addition to the first object point P1 on the object plane P, the processing unit 40 irradiates a second object point P2 with light of a first wavelength spectrum as first illumination light from the illumination unit 20 so that a bundle of light (luminous flux) forms a first solid angle A1 (step S12). A solid angle that does not include the first solid angle A1 is defined as a second solid angle A2. Then, the processing unit 40 causes the imaging unit 30 to capture an image of the first illumination light corresponding to the light of the first wavelength spectrum from the object P, thereby acquiring a first captured image (step S22).

[0077] The processing unit 40 causes the illumination unit 20 to irradiate the second object point P2 with light of the second wavelength spectrum as the second illumination light, simultaneously with the first illumination light, so that the light beam forms a third solid angle A3 and a fourth solid angle A4 at the second object point P2 (step S12). The third solid angle A3 is included in the first solid angle A1. At the same time, the fourth solid angle A4 is included in the second solid angle A2. Here, the third solid angle A3 and the fourth solid angle A4 are discontinuous regions. However, they do not have to be discontinuous and may be continuous. Then, the processing unit 40 causes the imaging unit 30 to capture the second illumination light from the object P and obtain a second captured image (step S22).

[0078] Light of the first wavelength spectrum of the first illumination light and light of the second wavelength spectrum of the second illumination light can be irradiated simultaneously, and are simultaneously dispersed and received by the imaging unit 30, thereby obtaining a first captured image corresponding to the light of the first wavelength spectrum of the first illumination light and a second captured image corresponding to the light of the second wavelength spectrum of the second illumination light.

[0079] Suppose a second object point P2 appears in the first captured image. In this case, the tilt angle θ can be identified as the region (range) of converted tilt angles converted from the first solid angle A1, and this region is designated as the first converted tilt angle region. On the other hand, suppose the second object point P2 does not appear in the first captured image. In this case, it is clear that the tilt angle θ is the estimated tilt angle of the region excluding the first converted tilt angle. The region including this estimated tilt angle of the second object point P2 is designated as the second converted tilt angle region. Therefore, the processing unit 40 acquires information about the object P at the second object point P2 (step S3).

[0080] Assume that a second object point P2 appears in the first captured image and also appears in the second captured image. In this case, the tilt angle can be identified as a region of converted tilt angles converted from the third solid angle A3, and this region is designated as the third converted tilt angle region. The third converted tilt angle region is narrower than the first converted tilt angle region. The angle at the second object point P2 in the third converted tilt angle region is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the second object point P2 (step S3).

[0081] Also, suppose that a second object point P2 appears in the first captured image, but does not appear in the second captured image. In this case, the tilt angle θ2 of the second object point P2 can be identified as the area obtained by excluding the third tilt angle area from the first converted tilt angle area. This is called the third converted tilt angle complementary area. This third converted tilt angle complementary area is narrower than the first converted tilt angle area. The angle at the second object point P2 in the third converted tilt angle complementary area becomes the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the second object point P2 (step S3).

[0082] On the other hand, suppose that the second object point P2 is not captured in the first captured image, but is captured in the second captured image. In this case, the tilt angle θ of the second object point P2 can be identified as a region of converted tilt angles converted from the fourth solid angle A4, and this region is designated as the fourth converted tilt angle region. This fourth converted tilt angle region is narrower than the second converted tilt angle region. The angle at the second object point P2 in the fourth converted tilt angle region is then the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the second object point P2 (step S3).

[0083] Also, suppose that the second object point P2 is not captured in the first captured image and is not captured in the second captured image. In this case, the tilt angle region θ2 of the second object point P2 can be identified as the region obtained by excluding the fourth tilt angle region from the second converted tilt angle region. This region is called the fourth converted tilt angle complementary region. This fourth converted tilt angle complementary region is narrower than the second converted tilt angle region. The angle at the second object point P2 in the fourth converted tilt angle complementary region becomes the estimated tilt angle. Therefore, the processing unit 40 acquires information about the object P at the second object point P2 (step S3).

[0084] As a result of the above, by using a combination of the first captured image and the second captured image, the processing unit 40 concludes that the inclination angle θ2 at the second object point P2 will become an estimated inclination angle in one of the four inclination angle regions: the third converted inclination angle region, the third converted inclination angle complementary region, the fourth converted inclination angle region, and the fourth converted inclination angle complementary region.

[0085] In the above example, the estimated tilt angle is calculated in step S3. In step S3, the converted tilt angle θ may be output in each loop other than the final loop. In this case, the estimated tilt angle may be calculated in step S3 of the final loop, for example.

[0086] The tilt angle θ2 obtained from only the first captured image can be in either the first converted tilt angle region or the second converted tilt angle region, depending on whether the second object point P2 is captured. Here, the first converted tilt angle region is larger than the third converted tilt angle region or the third converted tilt angle complementary region. The second converted tilt angle region is larger than the fourth converted tilt angle region or the fourth converted tilt angle complementary region. This means that the tilt angle θ2 (information about object P) obtained by combining the first and second captured images can be estimated with higher accuracy than the tilt angle θ2 (information about object P) obtained from only the first captured image.

[0087] On the other hand, the tilt angle θ2 obtained from only the second captured image can be a region that combines the third converted tilt angle region and the fourth converted tilt angle region, or a region that does not include them, depending on whether the second object point P2 is captured. The region that combines the third converted tilt angle region and the fourth converted tilt angle region is larger than the third converted tilt angle region or the fourth converted tilt angle region. Furthermore, the region that does not include the region that combines the third converted tilt angle region and the fourth converted tilt angle region is larger than the third converted tilt angle region or the fourth converted tilt angle region. This means that the tilt angle θ2 (information about object P) obtained by combining the first captured image and the second captured image can be estimated with higher accuracy than the tilt angle θ2 (information about object P) obtained from only the second captured image.

[0088] In this way, by combining the first captured image and the second captured image, the processing unit 40 can obtain more detailed information about the tilt angle θ2 of the object point P2 of the object P. That is, as in this embodiment, by defining the solid angle included in the first solid angle A1 as the third solid angle A3, defining the solid angle included in the second solid angle A2 as the fourth solid angle A4, and simultaneously forming ray bundles of the third solid angle A3 and the fourth solid angle A4 by the illumination unit 20, the accuracy of estimating the tilt angle θ2 can be improved. Furthermore, by inductively and repeatedly using the method of the optical inspection device 10 according to this embodiment, the accuracy of estimating the tilt angle θ2 can be improved.

[0089] In the above, we considered an example in which the object surface P is smooth and specularly reflects light. However, this is not limiting. The object surface P may also be a rough surface. In this case, the specular reflection component has a greater light intensity than the diffuse component. Then, when capturing the second object point P2, the processing unit 40 distinguishes whether the second object point P2 is captured in the captured image. However, here, the processing unit 40 may distinguish whether the light intensity (pixel value) of the second object point P2 is large or small. Therefore, even when the object surface P is a rough surface, the processing unit 40 can obtain a similar result by replacing the magnitude of the light intensity of the captured image of the second object point P2 with the magnitude of the light intensity of the captured image of the second object point P2. In other words, by combining the first captured image and the second captured image and comparing the magnitude of the light intensity of the captured image of the second object point P2, the estimation accuracy of the tilt angle θ2 of the object surface P can be improved. Therefore, this embodiment has the advantage of being able to acquire information about the shape of the object surface P having various BRDFs. Of course, the processing unit 40 may estimate the inclination angle θ of the object plane P as information about the object P by combining the first captured image and the second captured image, comparing the magnitude of the light intensity of the captured image of the second object point P2, and determining whether the second object point P2 is captured.

[0090] Furthermore, according to this embodiment, information on the object P at not only the first object point but also the second object point can be simultaneously acquired. In other words, this makes it possible to detect whether the state (shape or properties) of the second object point differs from the first object point. For example, it is possible to distinguish between the difference in surface inclination at the first object point and the second object point. Or, it is possible to distinguish between the difference in BRDF at the first object point and the second object point. For example, if the standard surface is a flat surface, the first object point is on the flat surface, and the second object point is on the flat surface but contains a minute defect, their BRDFs will be different. Therefore, it is possible to detect that the second point differs from the standard surface due to the difference in the intensity or wavelength spectrum of the reflected light from the two object points. In other words, it is possible to detect the presence or absence of a minute defect.

[0091] Therefore, according to this embodiment, it is possible to provide an optical inspection device 10, an optical inspection method, and an optical inspection program that are capable of acquiring information about an object P with illumination of a limited number of colors (wavelength spectrum).

[0092] (Third embodiment) The optical inspection device 10 according to this embodiment will be described below with reference to Fig. 7. The basic configuration of this embodiment is the same as that of the optical inspection device 10 according to the first embodiment. Therefore, only the differences will be described in this embodiment.

[0093] 7 is a schematic cross-sectional view of the optical inspection device 10 according to this embodiment. In the cross-sectional view shown in FIG. 7, the imaging optical axis C of the imaging unit 30 30 It is assumed that this includes:

[0094] The imaging unit 30 uses an image sensor 32 and an imaging optical element 34, and can form an image of light on the image sensor 32 by the imaging optical element 34. In this embodiment, the imaging optical element 34 is a lens combination. The imaging optical element 34 of the imaging unit 30 is called an imaging optical element.

[0095] The imaging unit 30 is located at the focal plane f of the imaging optical element 34. 30 The first light selector 36 is provided at the imaging optical axis C. The light selector 36 selectively passes light or changes the properties of light depending on the position where the light reaches. The light selector 36 may be, for example, an aperture. In this embodiment, the light selector 36 is 30 That is, the light selection unit 36 ​​is an aperture having a through hole on the optical axis C 30 The light selecting unit 36 ​​passes only light that reaches the upper part of the light axis C of the light selecting unit 36, and blocks other light. However, this is not a limitation, and the light selecting unit 36 ​​may be configured to change the intensity of light immediately after passing depending on the position where the light reaches, or to change the direction of light. For example, the light selecting unit 36 ​​may be configured to change the intensity of light immediately after passing depending on the polarization state of light at the position where the light reaches on the light selecting unit 36, or to change the intensity of light depending on the wavelength. The light selecting unit 36 ​​may be, for example, a wavelength filter having multiple regions that transmit different wavelength spectrums. In other words, for example, the light selecting unit 36 ​​may be configured to transmit light along the optical axis C of the light selecting unit 36. 30The light selecting unit 36 ​​may transmit light of a first wavelength spectrum that reaches the optical axis C and block light of a different wavelength spectrum. 30 Alternatively, the light of the first wavelength spectrum that has reached an area distant from the optical axis C of the light selection unit 36 ​​may be blocked. 30 Alternatively, the light selection unit 36 ​​may be a diffuser plate. In this case, the optical axis C of the diffuser plate may be set to 0.5 mm. 30 By providing a through-hole on top, only the light rays that pass through the through-hole pass through while maintaining their light ray direction. Therefore, the light rays that pass through in this way can form an image on the image sensor 32. On the other hand, the direction of the light rays that reach the area around the through-hole is changed by the diffuser plate. Therefore, an image of the object P cannot be formed on the image sensor 32. In other words, only the light rays that pass through the through-hole are imaged (captured).

[0096] Based on the above-described configuration, the operation of the optical inspection device 10 according to this embodiment will be described.

[0097] The processing unit 40 causes the illumination unit 20 to irradiate at least a first object point P1 on the object plane P with light of a first wavelength spectrum as first illumination light.

[0098] The focal plane f of the imaging optical element 34 30 Since the aperture 36 is disposed at the center of the image sensor 32 of the imaging unit 30, the light beam captured by the image sensor 32 is directed along the imaging optical axis C. 30 Only light rays parallel to the imaging optical axis C are captured. 30 Light rays traveling obliquely relative to the object P are blocked by the aperture 36. In this case, the imaging unit 30 is telecentric on the object side. As a result, even when the imaging unit 30 is brought close to the object P, there is a one-to-one relationship between the inclination angle θ of the smooth object surface P and the angle of incidence of light.

[0099] This also applies when at least the first object point P1 on the object plane P is irradiated with light of the first wavelength spectrum as the second illumination light.

[0100] This has the effect of enabling more detailed information about the tilt angle θ of the object P to be acquired by combining the first captured image and the second captured image. That is, as in this embodiment, the solid angle included in the first solid angle A1 is defined as the third solid angle A3, the solid angle included in the second solid angle A2 is defined as the fourth solid angle A4, and the illumination unit 20 simultaneously forms ray bundles of the third solid angle A3 and the fourth solid angle A4, thereby enabling the accuracy of estimating the tilt angle θ to be improved. Furthermore, by inductively and repeatedly using the method of the optical inspection device 10 according to this embodiment, the accuracy of estimating the tilt angle θ can be improved.

[0101] Therefore, according to this embodiment, it is possible to provide an optical inspection device 10, an optical inspection method, and an optical inspection program that are capable of acquiring information about an object P with illumination of a limited number of colors (wavelength spectrum).

[0102] (Variation) As the first light selecting unit 36 ​​of this embodiment, a wavelength filter having regions that selectively transmit two different wavelength spectra is used instead of an aperture. That is, the first light selecting unit 36 ​​of the modified example has an optical axis C of the light selecting unit 36. 30 The first light-selecting unit 36 ​​has a region with the top at the center that transmits light of a first wavelength spectrum that reaches the vicinity of the center, and a surrounding region that blocks light of the first wavelength spectrum and transmits light of a second wavelength spectrum that is different from the first wavelength spectrum. By forming the first light-selecting unit 36 ​​in this manner, light of the first wavelength spectrum becomes light of a wavelength spectrum different from the original when it passes through the surrounding region of the first light-selecting unit 36. Then, by acquiring an image using the imaging unit 30 equipped with the color image sensor 32, an imaging optical axis C 30 In this case, by distinguishing between the first wavelength spectrum and a spectrum different from the first wavelength spectrum, it is possible to obtain information about the object P similar to that obtained when the light selection unit 36 ​​is an aperture. Furthermore, the imaging optical axis C 30Since it is possible to obtain light rays oblique to the object P, it has the effect of obtaining more detailed information about the object P. This means that more detailed BRDF information can be obtained. Obtaining a detailed BRDF means that it is possible to obtain information about the shape and properties of the object's surface.

[0103] (Fourth embodiment) The optical inspection device 10 according to this embodiment will be described below with reference to Fig. 8. The basics of this embodiment are the same as those of the second embodiment (Fig. 6). Therefore, only the differences will be described in this embodiment.

[0104] FIG. 8 shows a schematic cross-sectional view of the optical inspection device 10 according to this embodiment. The optical inspection device 10 according to this embodiment includes an illumination unit 20, an imaging unit 30, and a beam splitter 50. The illumination of the object P and the imaging of the object P are performed via the beam splitter 50. By using the beam splitter 50, the illumination optical axis C 20 and imaging optical axis C 30 can be made optically coaxial.

[0105] In this embodiment, the illumination unit 20 includes a light source or a projection unit. In this embodiment, the illumination unit 20 includes a projection unit 22. The projection unit 22 can instantly change (vary) the projected image electrically. However, the projection unit 22 may also switch the projected image mechanically. In this embodiment, the projection unit 22 is a color projector and is a DLP (Digital Lighting Processing) equipped with a DMD (Digital Micromirror Device). However, this is not a limitation.

[0106] There are various types of projection unit 22. For example, a liquid crystal display (LCD) can be used to project images at enlarged, reduced, or normal magnification. These projection units 22 can instantly switch the projected image electrically. Alternatively, the projection unit 22 can instantly switch the projected image mechanically. For example, a slide projector that mechanically switches and projects slides, or an overhead projector (OHP) that projects an image drawn on a transparent sheet, may be used. Alternatively, the projection unit 22 may generate a projection pattern by combining a light source that can be electrically switched on and off with an imaging optical element. Alternatively, the projection unit 22 may generate a projection pattern by combining a color filter with an imaging optical element, a phosphor sheet or phosphor-coated plate with an imaging optical element, a dichroic mirror with an imaging optical element, or a polarizing plate with an imaging optical element.

[0107] The illumination unit 20 includes an illumination imaging optical element 24. The illumination imaging optical element 24 may be any imaging optical element, but in this example, it is assumed to be a lens assembly. The focal plane of the illumination imaging optical element 24 is referred to as an illumination focal plane f 20 Let's say.

[0108] The illumination unit 20 further includes a second light selection unit 26. In this embodiment, the light selection unit 26 is a transmission type diffusion plate. The projection unit 22 irradiates this diffusion plate 26 with a projection pattern.

[0109] Based on the above configuration, the operation of the optical inspection device 10 according to this embodiment will be described.

[0110] The processing unit 40 forms (images) a projection pattern on the transmission type diffusion plate 26 by the projection unit 22 provided in the illumination unit 20. The projection pattern is, for example, 20 Top and illumination optical axis C 20The diffuser 26 diffuses the light beams from the diffuser 26 to the object P, forming a first selected region 26a in the surrounding area. The first selected region 26a is formed, for example, in a circular shape using light of the first wavelength spectrum. However, this is not limited to this; the projection pattern may have any shape, any light intensity distribution, any color, multiple colors, any polarization, or anything else. The light then passes through the diffuser 26, is converted into light beams oriented in various directions, and is emitted toward the illumination imaging optical element 24. The light beams then reach substantially the entire surface of the lens, which is the illumination imaging optical element 24, provided in the illumination unit 20. This allows the illumination unit 20 to illuminate a wide area of ​​the object P. In other words, using such an illumination unit 20 widens the illumination field on the object P. Therefore, using the illumination unit 20 according to this embodiment, a bundle of light rays having a first solid angle A1 can be formed in an illumination field on a wide area of ​​the object P. On the other hand, without the diffuser plate 26, the light from the projection unit 22 would not reach the entire surface of the lens of the illumination imaging optical element 24, and the light beam having the first solid angle A1 would not be able to be formed over a wide range. In other words, the second light selection unit 26 has the effect of forming the light beam having the first solid angle A1 over a wide range.

[0111] The processing unit 40 uses the illumination unit 20 to form a bundle of rays at a first solid angle A1 simultaneously at at least the first object point P1 and the second object point P2 (step S11), and the imaging unit 30 acquires a first captured image (step S11). Next, similar to forming the bundle of rays at the first solid angle A1, the processing unit 40 changes (alters) the projection pattern of the projection unit 22 of the illumination unit 20 to form bundles of rays at a third solid angle A3 and a fourth solid angle A4 simultaneously at at least the first object point P1 and the second object point P2 (step S11), and the imaging unit 30 acquires a second captured image (step S21). As a result, the processing unit 40 can acquire information about the object plane P using the first captured image and the second captured image (step S3). Furthermore, in this embodiment, the projection pattern can be instantly changed by the projection unit 22. Therefore, the optical inspection device 10 according to this embodiment has the advantage of being able to instantaneously form a bundle of rays at the first solid angle A1 or the third solid angle A3 and the fourth solid angle A4, thereby obtaining information about the object P at a higher speed.

[0112] By using the illumination unit 20 according to this embodiment, a bundle of rays with the same solid angle can be formed simultaneously over a wide irradiation field on the object P. In other words, by using the illumination unit 20 according to this embodiment, it is possible to form a bundle of rays with the same solid angle simultaneously on at least the first object point P1 and the second object point P2. This has the effect that the optical inspection device 10 according to this embodiment can acquire information about the object P at higher speed. Furthermore, since it is possible to compare at least the first object point P1 and the second object point P2 simultaneously, it is possible to detect differences in the object information at each point.

[0113] Furthermore, by inductively and repeatedly using the method of the optical inspection device 10 according to this embodiment, the estimation accuracy of the tilt angle θ can be improved. For example, after acquiring the second captured image, a fifth solid angle and a sixth solid angle can be formed for the third solid angle A3, similar to the third solid angle A3 and the fourth solid angle A4 for the first solid angle A1. Then, a seventh solid angle and an eighth solid angle can be formed for the fourth solid angle A4, similar to the third solid angle A3 and the fourth solid angle A4 for the first solid angle A1. In this way, the number of solid angles can be exponentially increased, thereby improving the estimation accuracy of the tilt angle θ. In this embodiment, only the first wavelength spectrum, i.e., white light, is used. Even when only one color is used, the accuracy of acquiring information about the object P can be improved by inductively (or iteratively) using this embodiment.

[0114] Therefore, according to this embodiment, it is possible to provide an optical inspection device 10, an optical inspection method, and an optical inspection program that are capable of acquiring information about an object P with illumination of a limited number of colors (wavelength spectrum).

[0115] In the present embodiment, an example has been described in which the illumination unit 20 includes the diffuser plate 26. For example, if the light from the light source 22 can be made to reach the entire surface of the illumination imaging optical element 24, the diffuser plate 26 is not necessarily required.

[0116] (Variation 1) The projection unit 22 of the illumination unit 20 shown in FIG. 8 may be a color projector. In this case, the projection pattern can be formed using light of at least two different wavelength spectrums. These wavelength spectrums are referred to as a first wavelength spectrum and a second wavelength spectrum, respectively. The imaging unit 30, which includes a color image sensor 32, can then distinguish between and receive the two wavelength spectrums, thereby simultaneously acquiring a first captured image and a second captured image. This allows the optical inspection device 10 according to this first modification to acquire information about the object P more quickly.

[0117] Furthermore, by repeatedly and inductively applying the method of the optical inspection device 10 according to this embodiment, the processing unit 40 can improve the accuracy of estimating the tilt angle θ. For example, after acquiring the second captured image, the processing unit 40 can further form ray bundles of the first wavelength spectrum at the third solid angle A3, similar to the third solid angle A3 and fourth solid angle A4 relative to the first solid angle A1. Then, the processing unit 40 can form ray bundles of the second wavelength spectrum at the seventh and eighth solid angles relative to the fourth solid angle A4, similar to the third solid angle A3 and fourth solid angle A4 relative to the first solid angle A1. This exponentially increases the number of solid angles, thereby improving the accuracy of estimating the tilt angle. In this modification, only the first wavelength spectrum and the second wavelength spectrum, i.e., two colors, are used. Even if only two colors are used in this way, by using this modified example recursively (or iteratively), the optical inspection device 10 can improve the accuracy of acquiring information about the object P. However, the number of colors is not limited to two, and any number of colors can be used.

[0118] Therefore, the optical inspection device 10 according to this embodiment may acquire information about the object P according to the processing flow shown in FIG. 3, or may acquire information about the object P according to the processing flow shown in FIG.

[0119] (Variation 2) The second light-selecting unit 26 shown in FIG. 8 may be, for example, a thin transparent glass with a phosphor coated on its surface. Here, a phosphor that is excited by blue light and emits green light (first wavelength spectrum) is coated on the first selection region 26a. However, the light-selecting unit 26 may be any material that uses a phosphor (or fluorescent agent) to convert the wavelength of light. Furthermore, a blue laser light source is provided instead of the projection unit 22 of the illumination unit 20. This has the effect of allowing light of the first wavelength spectrum to reach the entire lens of the illumination imaging optical element 24 by irradiating the second selection unit 26 with blue laser light from the blue laser light source 22. Furthermore, while heat resistance is often taken into consideration for phosphors, in this modification, the phosphor is disposed away from the light source 22, which may be a heat source. This has the effect of improving the durability of the second light-selecting unit 26, which includes a phosphor.

[0120] (Variation 3) The second light selection section 26 shown in FIG. 8 may be a color filter (wavelength filter) having a transmission spectrum range that transmits the first wavelength spectrum in the first wavelength range.

[0121] The light selection unit 26 has an optical axis C of the light selection unit 26. 20 The light selecting unit 26 has an area centered on the top that transmits light of a first wavelength spectrum that reaches the vicinity of the center, and a surrounding area that blocks light of the first wavelength spectrum and transmits light of a second wavelength spectrum different from the first wavelength spectrum. 20 It may transmit light of a first wavelength spectrum that reaches it and block light of a different wavelength spectrum.

[0122] A white LED is provided in place of the projection unit 22 of the illumination unit 20. This has the effect of easily forming a bundle of rays at the first solid angle A1.

[0123] (Variation 4) There are various projection patterns by the projection unit 22 described in the fourth embodiment (see FIG. 8). One example is shown in FIG. 9. FIG. 9 shows the projection pattern of the illumination optical axis C20 The illumination focal plane f is perpendicular to 20 9A shows examples of projection patterns projected onto the first selected area 26a from the projection unit 22. The example shown in FIG. 9A shows the projection patterns projected onto the focal plane f of the illumination imaging optical element 24 from the projection unit 22. 20 9B is a first projection pattern Pa1 projected toward the second light selection unit 26. The example shown in FIG. 9B is a first projection pattern Pa1 projected from the projection unit 22 to the focal plane f of the illumination imaging optical element 24. 20 10 shows a second projection pattern Pa2 projected toward the second light selection unit 26.

[0124] The projection unit 22 of the illumination unit 20 projects the first projection pattern Pa1 onto the focal plane f of the illumination imaging optical element 24. 20 The light beams are projected toward the second light selection unit 26, forming a bundle of rays at a first solid angle A1 at least at the first object point P1 and the second object point P2 simultaneously (step S11), and the imaging unit 30 acquires a first captured image (step S11). Next, in the same manner as when the bundle of rays at the first solid angle A1 is formed, the projection pattern of the projection unit 22 is changed (modified) to project the second projection pattern Pa2 onto the focal plane f of the illumination imaging optical element 24. 20 The light beams are projected toward the second light selection unit 26, forming bundles of rays at a third solid angle A3 and a fourth solid angle A4 simultaneously at at least the first object point P1 and the second object point P2 (step S11), and the imaging unit 30 acquires a second captured image (step S11). As a result, the processing unit 40 can acquire information about the object surface P using the first captured image and the second captured image (step S3).

[0125] The illumination unit 20 of this modified example can instantaneously change the projection patterns Pa1 and Pa2 using the projection unit 22. This has the effect of enabling the illumination unit 20 to instantaneously form a bundle of rays at the first solid angle A1 or the third solid angle A3 and the fourth solid angle A4. This has the effect of enabling the optical inspection device 10 of this modified example to acquire information about the object P at higher speed.

[0126] With this modification, the illumination optical axis C 20 For example, an axially symmetric solid angle can be formed with respect to the illumination optical axis C as the inclination angle of the inclined surface θ of the object P. 20This has the effect of enabling angle information to be obtained.

[0127] (Variation 5) The projection pattern by the projection unit 22 described in the fourth embodiment (see FIG. 8) can be various other than the fourth modification. One example is shown in FIG. 10. FIG. 10 shows the projection pattern of the illumination optical axis C 20 The illumination focal plane f is perpendicular to 20 10A shows examples of projection patterns projected onto the first selected area 26a from the projection unit 22. The example shown in FIG. 10A shows the projection patterns projected onto the focal plane f of the illumination imaging optical element 24 from the projection unit 22. 20 10B is a first projection pattern Pa1 projected toward the second light selection unit 26. The example shown in FIG. 10B is a first projection pattern Pa1 projected from the projection unit 22 to the focal plane f of the illumination imaging optical element 24. 20 10 shows a second projection pattern Pa2 projected toward the second light selection unit 26.

[0128] The projection unit 22 of the illumination unit 20 projects the first projection pattern Pa1 onto the focal plane f of the illumination imaging optical element 24. 20 The light beams are projected toward the second light selection unit 26, forming a bundle of rays at a first solid angle A1 at at least the first object point P1 and the second object point P2 simultaneously (step S11), and the imaging unit 30 acquires a first captured image (step S21). Next, in the same manner as when the bundle of rays at the first solid angle A1 is formed, the projection pattern of the projection unit 22 is changed (modified) to project the second projection pattern Pa2 onto the focal plane f 20 The light beam is projected toward the object plane P, forming a bundle of rays at a third solid angle A3 and a fourth solid angle A4 simultaneously at at least the first object point P1 and the second object point P2 (step S11), and the imaging unit 30 acquires a second captured image (step S21). This allows information about the object plane P to be acquired using the first captured image and the second captured image (step S3).

[0129] The illumination unit 20 of this modified example can instantaneously change the projection patterns Pa1 and Pa2 using the projection unit 22. This has the effect of enabling the illumination unit 20 to instantaneously form a bundle of rays at the first solid angle A1 or the third solid angle A3 and the fourth solid angle A4. This has the effect of enabling the optical inspection device 10 of this modified example to acquire information about the object P at higher speed.

[0130] In this modification, since the projection patterns Pa1 and Pa2 have translational symmetry in one axis direction, there is an advantage that angle information with respect to a direction perpendicular to the translational symmetry direction can be obtained as the tilt angle θ of the tilted surface of the object P.

[0131] (Variation 6) The projection pattern by the projection unit described in the fourth embodiment (see FIG. 8) can be various other than the modified examples 4 and 5. One example is shown in FIG. 11. FIG. 11 shows the projection pattern of the illumination optical axis C 20 The illumination focal plane f is perpendicular to 20 11A shows examples of projection patterns projected onto the first selected area 26a from the projection unit 22. The example shown in FIG. 11A shows the projection patterns projected onto the focal plane f of the illumination imaging optical element 24 from the projection unit 22. 20 11B is a first projection pattern Pa1 projected toward the second light selection unit 26. The example shown in FIG. 11B is a first projection pattern Pa1 projected from the projection unit 22 to the focal plane f of the illumination imaging optical element 24. 20 10 shows a second projection pattern Pa2 projected toward the second light selection unit 26.

[0132] The projection unit 22 of the illumination unit 20 projects the first projection pattern Pa1 onto the focal plane f of the illumination imaging optical element 24. 20 The light beams are projected toward the second light selection unit 26, forming a bundle of rays at a first solid angle A1 at at least the first object point P1 and the second object point P2 simultaneously (step S11), and the imaging unit 30 acquires a first captured image (step S21). Next, in the same manner as when the bundle of rays at the first solid angle A1 is formed, the projection pattern of the projection unit 22 is changed (modified) to project the second projection pattern Pa2 onto the focal plane f 20The light beam is projected toward the object plane P, forming a bundle of rays at a third solid angle A3 and a fourth solid angle A4 simultaneously at at least the first object point P1 and the second object point P2 (step S11), and the imaging unit 30 acquires a second captured image (step S21). This allows information about the object plane P to be acquired using the first captured image and the second captured image (step S3).

[0133] The illumination unit 20 of this modified example can instantaneously change the projection patterns Pa1 and Pa2 using the projection unit 22. This has the effect of enabling the illumination unit 20 to instantaneously form a bundle of rays at the first solid angle A1 or the third solid angle A1 and the fourth solid angle A4. This has the effect of enabling the optical inspection device 10 of this modified example to acquire information about the object P at higher speed.

[0134] In this modified example, since the projection patterns Pa1 and Pa2 have a distribution in the azimuth direction, there is an advantage that angle information relative to the azimuth direction can be obtained as the tilt angle θ of the tilted surface of the object P.

[0135] (Variation 7) The projection patterns by the projection unit 22 described in the fourth embodiment (see FIG. 8) include various other patterns in addition to the modified examples 4, 5, and 6. One example is shown in FIG. 12. FIG. 12 shows the projection patterns of the illumination optical axis C 20 The illumination focal plane f is perpendicular to 20 The image above represents a projection pattern projected onto the first selected area 26a from the projection unit 22. Here, the first wavelength spectrum is blue light with a peak wavelength of 450 nm and a wavelength width of 100 nm. The second wavelength spectrum is red light with a peak wavelength of 650 nm and a wavelength width of 100 nm.

[0136] The projection unit 22 of the illumination unit 20 projects the first projection pattern Pa1 onto the focal plane f of the illumination imaging optical element 24. 20The light is projected toward the second light selection unit 26a1 and simultaneously irradiated toward at least the first object point P1 and the second object point P2 (step S12). Here, the first selected region 26a1 is formed with light of a first wavelength spectrum, and the third selected region 26a3 and the fourth selected region 26a4 are formed with light of a second wavelength spectrum. The light from the first selected region 26a1 forms a bundle of rays at a first solid angle A1, and the light from the third selected region 26a3 and the fourth selected region 26a4 form bundles of rays at a third solid angle A3 and a fourth solid angle A4. Then, a captured image is acquired by the imaging unit 30 including the color image sensor 32 (step S22). The color image sensor 32 is assumed to be able to distinguish between blue light and red light. The image captured using blue light is assumed to be, for example, a first captured image, and the image captured using red light is assumed to be, for example, a second captured image. As a result, information about the object surface P can be acquired using the first captured image and the second captured image (step S3). In this embodiment, the first captured image and the second captured image can be acquired simultaneously without instantaneously changing the projection pattern by the projection unit 22. This has the effect of enabling information about the object P to be acquired at high speed.

[0137] In addition to the projection patterns described above, there are various other projection patterns that can be produced by the projection unit. That is, any projection pattern can be used as long as it can simultaneously or chronologically form ray bundles at the first solid angle A1, the third solid angle A3, and the fourth solid angle A4. However, if they are simultaneously formed, the ray bundles at the third solid angle A3 and the fourth solid angle A4 must be distinguishable from the ray bundle at the first solid angle A1 by wavelength, polarization, intensity, etc. As long as this requirement is met, the shape of the projection pattern, the color of light (wavelength spectrum), the polarization of light, the intensity, or the intensity distribution of light can be any. Furthermore, the ray bundles at the third solid angle A4 and the fourth solid angle A4 may have different wavelengths, polarization, and intensities.

[0138] (Application example 1) Hereinafter, an application example 1 of the optical inspection device 10 according to this embodiment will be described with reference to Fig. 13. The optical inspection device 10 used in this application example 1 is the same as the optical inspection device 10 described in the fourth embodiment (Fig. 8).

[0139] In this application example 1, the projection pattern is projected onto the illumination focal plane f 20 An algorithm for repeatedly obtaining the tilt angle θ of the object plane P by acquiring a captured image each time the object plane P is formed and adding up the increment of the tilt angle θ will be described below.

[0140] The operation of the optical inspection device 10 in Application Example 1 will be described.

[0141] As shown in Figure 13(a), the surface of object P is a reflective convex surface. The convex surface is sufficiently smooth, and reflectivity means that incident light is specularly reflected. In Figure 13(a), an approximately circular cone with a height of approximately 50 μm and a diameter of approximately 2000 μm is drawn, color-coded according to the inclination angle. The inclination angle at the point where the approximate cone connects to the base and at the apex is 0°.

[0142] The processing unit 40 receives the light from the projection unit 22 of the illumination unit 20 at the focal plane f 20 13(e) are projected onto the projection pattern Pa1. Here, white light is assumed to be projected. Therefore, the processing unit 40 projects the projection patterns Pa1 and Pa2 in sequence. The first projection pattern Pa1 is one ring. On the other hand, the second projection pattern Pa2 is two rings. In other words, the second projection pattern Pa2 has twice as many rings as the first projection pattern Pa1. Note that the first projection pattern Pa1 is the same as that shown in FIG. 9(A), and the second projection pattern Pa2 is the same as that shown in FIG. 9(B).

[0143] In the second projection pattern Pa2, white light is not projected between the portion of one of the rings of the first pattern Pa2 indicated by the dashed line and its inner edge. Also, in the second projection pattern Pa2, the inner edge of one of the rings of the first pattern Pa1 is divided into two in the radial direction, and white light is projected onto the outer ring, but not onto the inner circle.

[0144] First, the processing unit 40 projects the first projection pattern Pa1 onto the focal plane f of the illumination imaging optical element 24. 20The object P is then imaged by the imaging unit 30, and a bundle of rays at a first solid angle A1 is projected onto the object surface P (step S11), and the object P is imaged by the imaging unit 30 to obtain a first image (step S21). A solid angle excluding the first solid angle A1 is defined as a second solid angle A2.

[0145] Next, the processing unit 40 projects the second projection pattern Pa2 onto the focal plane f of the illumination imaging optical element 24. 20 The light beams at the third solid angle A3 and the fourth solid angle A4 are projected onto the object surface P (step S11), and the object P is imaged by the imaging unit 30 to obtain a second image (step S21). Here, the third solid angle A3 is included in the first solid angle A1. The fourth solid angle A4 is included in the second solid angle A2. The light beams at the third solid angle A3 and the fourth solid angle A4 are simultaneously formed by the illumination unit 20.

[0146] 13(c) shows the first captured image Ic1 and the second captured image Ic2 captured by the imaging unit 30. In the first captured image Ic1 and the second captured image Ic2, areas with high light intensity are white, and areas with low light intensity are black.

[0147] In the first captured image Ic1, the black area has a converted tilt angle of 0.0° to 3.0°. The white area has a converted tilt angle of 3.0° to 6.0°. Therefore, the angular resolution here is 3.0°. In this algorithm, the estimated tilt angle of the white area in the first captured image Ic1 is set to 3.0°, the minimum converted tilt angle. This distribution is referred to as the first estimated tilt angle distribution and is shown by symbol Ib1 in Figure 13(b).

[0148] In the second captured image Ic2, the regions appear as black, white, black, white, and black, in that order from the outside to the inside (vertex). The angular resolution here is 1.5°. In the second captured image Ic2, the black regions have converted tilt angles of 0.0° to 1.5° or 3.0° to 4.5°. The white regions have converted tilt angles of 1.5° to 3.0° or 4.5° to 6.0°. In this algorithm, for the tilt angles of regions that are black in the first captured image and white in the second captured image, the first estimated tilt angle is added by 1.5°, the angle of the angular resolution. In addition, calculations are performed to maintain the tilt angles of regions that are white in the first captured image and black in the second captured image. By performing these calculations, a shape that reflects the angles of the areas where the object points are captured and those that are not captured is created.

[0149] Although not shown, the third pattern light Pa3 forms a solid angle relative to the second pattern light Pa2 as described above, and a third captured image is acquired. The angular resolution at this time is 0.75°. In this algorithm, for the inclination angle of a region where the second captured image is black and the third captured image is white, 0.75°, the angle of the angular resolution, is added to the estimated inclination angle. Furthermore, a calculation is performed to maintain the inclination angle of a region where the second captured image is white and the third captured image is black. By performing this calculation, a shape that reflects the angle is created for the areas where the object points are captured and those where they are not captured.

[0150] The same process is repeated thereafter, doubling the number of rings in the projection pattern. That is, the number of rings in the third projection pattern is set to four, and the number of rings in the fourth projection pattern is set to eight. Then, captured images are acquired for each. This has the effect of improving the estimation accuracy of the tilt angle distribution θ with each iteration. In other words, the effect is that the gradation of the angular resolution increases exponentially: 2x, 4x, 8x, 16x.

[0151] Here, the estimated tilt angle distribution obtained by combining the first captured image Ic1 and the second captured image Ic2 has higher estimation accuracy than the estimated tilt angle distribution obtained from the first captured image Ic1. In other words, increasing the number of captured images and combining them has the effect of increasing estimation accuracy.

[0152] Here, the estimated tilt angle distribution using the first captured image Ic1 and the second captured image Ic2 has an angle range of 0° to 6° and an angular resolution of 1.5°. In other words, the angular resolution is four levels. If the number of rings in the second projection pattern Pa2 were not twice that of the first projection pattern Pa1—for example, if the second projection pattern Pa2 had only one ring—the first captured image Ic1 and the second captured image Ic2 would only achieve a maximum angular resolution of three levels. In other words, as in this embodiment, the solid angle included in the first solid angle A1 is defined as the third solid angle A3, the solid angle included in the second solid angle A2 is defined as the fourth solid angle A4, and the illumination unit 20 simultaneously forms ray bundles of the third solid angle A3 and the fourth solid angle A4, thereby improving the estimation accuracy of the tilt angle θ.

[0153] (Application example 2) Fig. 14(a) shows an example in which an algorithm identical to that of Application Example 1 described above is applied to a somewhat more complex uneven surface, in contrast to the convex surface shown in Fig. 13(a). Object P shown in Fig. 14(a) is an example in which a convex cone with a height of, for example, approximately 50 µm from the plane and a diameter of approximately 2000 µm is adjacent to a concave cone with a depth of, for example, approximately 50 µm from the plane and a diameter of approximately 2000 µm. That is, Fig. 14(b) shows an example in which the surface of object P is a combination of reflective convex and concave surfaces. The convex and concave surfaces are sufficiently smooth, and the reflectivity means that incident light is specularly reflected.

[0154] In this example, the first projection pattern Pa1 shown in FIG. 13(e), the second projection pattern Pa2, and the third projection pattern (not shown), etc., were projected in sequence, and the first captured image, the second captured image, the third captured image, etc., shown in FIG. 14(c), were obtained, respectively.

[0155] Then, the processing unit 40 obtained a first estimated tilt angle distribution Ib1, a second estimated tilt angle distribution Ib2, and a third estimated tilt angle distribution Ib3 from these captured images, as shown in Figure 14(b), while adding angular resolution appropriately depending on whether or not an object point was captured on the image.

[0156] The processing unit 40 then repeats the process to increase the accuracy of the tilt angle distribution. In this way, in Application Example 2, information (tilt angle θ) of each object point of the object P can be obtained.

[0157] Although not shown in detail, for example, the captured image shown in the upper part of Fig. 14(c') can be obtained using an appropriate pattern light, and the captured image shown in the lower part of Fig. 14(c') can be obtained using a more finely divided pattern light. Then, the first estimated tilt angle distribution Ib'1 and the second estimated tilt angle distribution Ib'2 shown in Fig. 14(b') can be obtained.

[0158] In this example, too, the accuracy of estimating the tilt angle distribution θ improves with each iteration. In other words, the angular resolution scale increases exponentially, for example, by 2x, 4x, 8x, and 16x. In this way, by repeatedly using this algorithm, the tilt angle distribution of various surfaces can be obtained with high accuracy.

[0159] (Application example 3) The object surface P is a collection of object points. It is known that once the tilt angle distribution θ of each object point of the object P is determined, the three-dimensional shape of the object surface P can be reconstructed using a deep neural network (see Non-Patent Document 3). For example, FIG. 15 shows the three-dimensional shape of the convex surface shown in FIG. 13, which is iteratively reconstructed using the tilt angles θ estimated five times. Therefore, by acquiring the tilt angle θ for each object point, the processing unit 40 of the optical inspection device 10 according to this embodiment can measure the shape of the object P, as shown in the bottom part of FIG. 15(c) (the same as FIG. 13(a)). In this way, there is an effect that the three-dimensional shape of the object P can be acquired.

[0160] 14(a), a shape having convex and concave portions relative to a plane can also be reconstructed using a deep neural network. Therefore, the processing unit 40 of the optical inspection device 10 according to this embodiment can measure the shape of the object P even if the shape is appropriately complex.

[0161] Although an example of measuring the shape of the object P has been described here, the optical inspection device 10, the optical inspection method, and the optical inspection program may also be used to inspect the surface or inner surface of the object P.

[0162] According to at least one of the embodiments described above, it is possible to provide an optical inspection device 10, an optical inspection method, and an optical inspection program that are capable of acquiring information about an object P with illumination of a limited number of colors (wavelength spectrum).

[0163] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0164] 10...optical inspection device, 20...illumination unit, 22...light source (projection unit), 24...illumination imaging optical element, 26...light selection unit, 26a...selection area, 30...imaging unit, 32...image sensor, 34...imaging imaging optical element, 36...light selection unit, 40...processing unit, 50...beam splitter, A1...first solid angle, A2...second solid angle, A3...third solid angle, A4...fourth solid angle, C 20 …Lighting optical axis, C 30 …imaging optical axis, f 20 …Illumination focal plane, f 30 ...imaging focal plane, P1...first object point, P2...second object point.

Claims

1. an illumination unit that irradiates one or more beams of light at a solid angle onto at least a first object point of an object; an imaging unit that acquires an image of the object in response to illumination by the bundle of rays of one or more solid angles; a light beam of a first solid angle formed by a first illumination light from the illumination unit is irradiated onto the first object point of the object, and a solid angle that does not include the first solid angle is defined as a second solid angle; acquiring a first captured image of the object using the imaging unit based on illumination by the first illumination light; a solid angle included in the first solid angle is defined as a third solid angle, a solid angle included in the second solid angle is defined as a fourth solid angle, and a bundle of rays of the second illumination light from the illumination unit at the third solid angle and the fourth solid angle is irradiated onto at least the first object point; acquiring a second captured image of the object using the imaging unit based on illumination by the second illumination light; a processing unit that acquires information about the object based on the first captured image and the second captured image; An optical inspection device comprising:

2. the processing unit irradiates at least the first object point of the object with a ray bundle of the first wavelength spectrum at the first solid angle as the first illumination light of the illumination unit, and before or after that, irradiates at least the first object point with a ray bundle of the first wavelength spectrum at the third solid angle and the fourth solid angle as the second illumination light of the illumination unit; The optical inspection device according to claim 1 .

3. the processing unit irradiates at least the first object point of the object with a ray bundle of a first wavelength spectrum at the first solid angle as the first illumination light of the illumination unit, and simultaneously irradiates at least the first object point with a ray bundle of a second wavelength spectrum different from the first wavelength spectrum at the third solid angle and the fourth solid angle as the second illumination light of the illumination unit. The optical inspection device according to claim 1 .

4. When an object point of the object that is different from the first object point is set as a second object point, the processing unit: causing a bundle of rays of light within the first solid angle to be irradiated onto the second object point in addition to the first object point of the object by the first illumination light from the illumination unit; acquiring the first captured image of the object using the imaging unit based on illumination by the first illumination light; causing a bundle of rays of light at the third solid angle and the fourth solid angle to be irradiated onto the second object point in addition to the first object point by the second illumination light from the illumination unit; acquiring the second captured image of the object using the imaging unit; acquiring information about the object based on the first captured image and the second captured image; 3. The optical inspection device according to claim 1 or 2.

5. the imaging unit includes an imaging optical element and a first light selection unit that controls passage of light to a focal plane of the imaging optical element; 3. The optical inspection device according to claim 1 or 2.

6. the illumination unit includes an illumination imaging optical element and a second light selection unit that controls passage of light to a focal plane of the illumination imaging optical element; 3. The optical inspection device according to claim 1 or 2.

7. the second light-selecting portion is axially symmetric; The optical inspection device according to claim 6 .

8. the second light selection unit has translational symmetry in one axial direction; The optical inspection device according to claim 6 .

9. the second light selection unit has a distribution in the azimuth angle direction; The optical inspection device according to claim 6 .

10. The processing unit measures the shape of the object.

3. The optical inspection device according to claim 1 or 2.

11. The processing unit inspects a surface or an inner surface of the object.

3. The optical inspection device according to claim 1 or 2.

12. irradiating at least a first object point of the object with a bundle of rays of a first solid angle by a first illumination light from an illumination unit, wherein a solid angle not including the first solid angle is defined as a second solid angle; acquiring a first captured image of the object using an imaging unit based on illumination by the first illumination light; defining a solid angle included in the first solid angle as a third solid angle, defining a solid angle included in the second solid angle as a fourth solid angle, and irradiating at least the first object point with ray bundles of the second illumination light from the illumination unit at the third solid angle and the fourth solid angle; acquiring a second captured image of the object using the imaging unit based on illumination by the second illumination light; acquiring information about the object based on the first captured image and the second captured image; An optical inspection method comprising:

13. When an object point of the object different from the first object point is defined as a second object point, illuminating the bundle of rays of the first solid angle at at least the first object point of the object includes illuminating the bundle of rays of the first solid angle at the second object point in addition to the first object point; irradiating at least the first object point with the ray bundles of the third solid angle and the fourth solid angle includes irradiating the second object point in addition to the first object point. The optical inspection method according to claim 12.

14. irradiating the light beam at the first solid angle by the first illumination light from the illumination unit includes irradiating a light beam of a first wavelength spectrum; irradiating the light beams at the third solid angle and the fourth solid angle by the second illumination light from the illumination unit includes irradiating the light beams of the first wavelength spectrum; The first illumination light and the second illumination light are irradiated at different times. The optical inspection method according to claim 12.

15. the illumination method includes simultaneously illuminating at least the first object point of the object with a bundle of rays of a first wavelength spectrum at the first solid angle as the first illumination light, and illuminating at least the first object point with bundles of rays of a second wavelength spectrum different from the first wavelength spectrum at the third solid angle and the fourth solid angle as the second illumination light. The optical inspection method according to claim 12 or 13.

16. A fifth solid angle and a sixth solid angle are formed relative to the third solid angle in the same relationship as the third solid angle and the fourth solid angle relative to the first solid angle; forming a seventh solid angle and an eighth solid angle with respect to the fourth solid angle in the same relationship as the third solid angle with respect to the first solid angle, and irradiating at least the first object point of the object with ray bundles of the third illumination light at the fifth solid angle, the sixth solid angle, the seventh solid angle, and the eighth solid angle; acquiring a third captured image of the object using the imaging unit based on illumination by the third illumination light; Including, acquiring information about the object based on the first captured image and the second captured image includes acquiring information about the object based on the first captured image, the second captured image, and the third captured image. The optical inspection method according to claim 12 or 13.

17. a light beam bundle of a first illumination light from an illumination unit is irradiated onto at least a first object point of an object within a first solid angle, wherein a solid angle not including the first solid angle is defined as a second solid angle; acquiring a first captured image of the object using an imaging unit based on illumination by the first illumination light; defining a solid angle included in the first solid angle as a third solid angle, defining a solid angle included in the second solid angle as a fourth solid angle, and irradiating at least the first object point with ray bundles of the second illumination light from the illumination unit at the third solid angle and the fourth solid angle; acquiring a second captured image of the object using the imaging unit based on illumination by the second illumination light; acquiring information about the object based on the first captured image and the second captured image; An optical inspection program that causes a computer to execute the above.

Images