JP2020139821A - Inspection device, inspection system and inspection method - Google Patents

Abstract

To provide a technology that improves estimation accuracy of a surface roughness parameter of an inspection object.SOLUTION: One aspect of the present invention is an inspection device that comprises: an irradiation unit that includes a plurality of light source units being located in a prescribed plane, and radiating light of a linear polarization in a prescribed polarization direction; an imaging unit that, with the light radiated by the irradiation unit and subjected to a mirror-surface reflection, diffuse reflection or sub-surface scattering upon an inspection object as secondary light, receives a polarization which is extracted by a polarization filter, in a plurality of polarization directions of the secondary light, and creates, for each polarization direction, image group information serving as a group of image data on a two-dimensional image for each of a plurality of light wavelengths predetermined on the basis of intensity of the received polarization; a separation unit that acquires a wavelength spectrum of the mirror-surface reflection light on the basis of a plurality of image group information created by the imaging device; and a surface roughness calculation unit that calculates a surface roughness parameter on the basis of the wavelength spectrum acquired by the separation unit.SELECTED DRAWING: Figure 2

Description

The present invention relates to an inspection device, an inspection system and an inspection method.

In order to inspect the appearance of an inspection object having a complicated three-dimensional shape such as a tire, a technique for measuring surface irregularities with high accuracy using polarized light has been proposed (see Patent Document 1). In the proposed technique, the surface of the inspection target is linearly irradiated with polarized light from a single direction, and the surface roughness parameter is calculated by analyzing the polarization state of the reflected light. However, since the irradiation is performed from a single direction, the accuracy of the surface roughness parameter may be poor in the proposed technique.
Therefore, an inspection technique for irradiating an inspection target with a planar polarized light source instead of the linear polarized light source has been proposed (see Patent Document 2). In the proposed technique, the surface roughness parameter is calculated by analyzing an image taken by a camera of the specular reflection of a planar light source (see Non-Patent Document 1 and Non-Patent Document 2). However, when the inspection target has a three-dimensional shape, the camera may not be able to receive specular reflection, and in such a case, the accuracy of the surface roughness parameter may be poor.

Japanese Unexamined Patent Publication No. 2011-196741 Japanese Unexamined Patent Publication No. 2013-213836

David A. Forsyth, "Computer Vision: A Modern Approach Second Edition", Pearson Publishing, 2011 Zhengyou Zhang, "Flexible Camera Calibration by Viewing a Plane from Unknown Orientations", IEEE, 0-7695-0164-8 / 99, 1999

In addition to the above problems, neither of the techniques disclosed in Patent Document 1 and Patent Document 2 is a technique considering the influence of subsurface scattering. Therefore, conventionally, the estimation accuracy of the surface roughness parameter may be poor depending on the inspection target.

In view of the above circumstances, it is an object of the present invention to provide a technique for improving the estimation accuracy of the surface roughness parameter to be inspected.

One aspect of the present invention is that the light emitted by the irradiation unit including a plurality of light source units located in a predetermined plane and emitting linearly polarized light in a predetermined polarization direction is specularly reflected, diffusely reflected, or diffused depending on the inspection target. The light scattered under the surface is used as the secondary light, and the polarized light extracted by the polarizing filter, which receives the polarized light in a plurality of polarization directions of the secondary light, is predetermined based on the intensity of the received polarized light. Based on an image pickup unit that generates image set information, which is a set of image data of two-dimensional images for each of a plurality of light wavelengths, for each of the polarization directions, and a plurality of the image set information generated by the image pickup unit. A separation unit that acquires the wavelength spectrum of specularly reflected light, and a surface roughness calculation unit that calculates a surface roughness parameter representing the surface roughness of the inspection target based on the wavelength spectrum acquired by the separation unit. It is an inspection device to be equipped.

One aspect of the present invention includes an irradiation unit including a plurality of light source units located in a predetermined plane and emitting linearly polarized light in a predetermined polarization direction, a polarizing filter for extracting polarized light in a predetermined polarization direction, and a polarizing filter. The light emitted by the irradiation unit is mirror-reflected, diffusely reflected, or subsurfacely scattered by the inspection target, and is polarized light extracted by the polarizing filter, with the light being scattered under the surface as the secondary light, and is a plurality of the secondary light. Imaging that receives polarized light in the polarization direction and generates image set information that is a set of image data of two-dimensional images for each of a plurality of predetermined light wavelengths based on the intensity of the received polarized light for each polarization direction. A separation unit that acquires a wavelength spectrum of mirror-reflected light based on the unit, a plurality of image set information generated by the imaging unit, and a surface of the inspection target based on the wavelength spectrum acquired by the separation unit. It is an inspection system including a surface roughness calculation unit for calculating a surface roughness parameter representing roughness.

One aspect of the present invention is an irradiation step in which an irradiation unit including a plurality of light source units located in a predetermined plane and emitting linearly polarized light in a predetermined polarization direction irradiates the light, and the irradiation unit irradiates the light. The light is mirror-reflected, diffusely reflected, or subsurfacely scattered by the inspection target, and is used as secondary light, which is polarized light extracted by a polarizing filter and receives polarized light in a plurality of polarization directions among the secondary light. An imaging step of generating image set information, which is a set of image data of a plurality of predetermined two-dimensional images for each light wavelength based on the intensity of the received polarized light, for each polarization direction, and an imaging step. The separation step of acquiring the wavelength spectrum of the mirror-reflected light based on the plurality of the image set information generated in the above step, and the surface roughness of the inspection target are represented based on the wavelength spectrum acquired in the separation step. It is an inspection method including a surface roughness calculation step for calculating a surface roughness parameter.

According to the present invention, it is possible to improve the estimation accuracy of the surface roughness parameter to be inspected.

The figure which shows an example of the functional structure of the inspection system 1 of embodiment. Explanatory drawing explaining the outline of the principle which concerns on inspection system 1 of embodiment. The figure which shows an example of the functional structure of the detection part 51 in an embodiment. The flowchart which shows an example of the process which the inspection system 1 of embodiment executes. The figure which shows an example of the experimental result of the inspection target image in an embodiment. The figure which shows an example of the functional structure of the inspection system 1a in the modification. The figure which shows an example of the restoration part 516 in the modification. The figure which shows an example of the experimental result which shows the correspondence relation between the change of the angle β formed in the modification, and the change of a specular reflection component. The figure which shows an example of the functional structure of the inspection system 1b of a modification. The figure which shows an example of the functional structure of the detection part 51b in the modification. The flowchart which shows an example of the process which the inspection system 1b of a modification is executed.

FIG. 1 is a diagram showing an example of the functional configuration of the inspection system 1 of the embodiment.
The inspection system 1 includes an irradiation unit 10, a polarizing filter 20, an image pickup device 40, an inspection device 50, and an inspection system control unit 60.

The irradiation unit 10 includes a plurality of light source units 101. The plurality of light source units 101 are located in a predetermined plane. The light emitted by the irradiation unit 10 is linearly polarized light in a predetermined polarization direction. The predetermined polarization direction may be, for example, a polarization direction parallel to a predetermined surface on which the light source unit 101 is located. The light emitted by the irradiation unit 10 is incident on the inspection target 9 at an incident angle corresponding to the position of each light source unit 101. The light emitted by the irradiation unit 10 to the inspection target 9 is light emitted by a part or all of the plurality of light source units 101. The plurality of light source units 101 may be arranged in any manner, and may be arranged in a grid pattern in a predetermined plane, for example. For example, if the shape of the plurality of light source units 101 is rod-shaped, the plurality of light source units 101 may be arranged in a row in a predetermined plane in a direction perpendicular to the rod. Further, in such a case, the predetermined polarization direction may be, for example, a direction parallel to the rod or a direction perpendicular to the rod.

The irradiation unit 10 is, for example, a planar polarized light source having a display function. A planar polarized light source having a display function is a display that displays an image of an arbitrary shape. The display image displayed by the irradiation unit 10 may be a linear image or a dot image. The display image displayed by the irradiation unit 10 may be a plurality of figures displayed at the same time.
The display function included in the irradiation unit 10 is realized by the radiation of light by a part or all of the plurality of light source units 101.
The display of the display image by the irradiation unit 10 is an operation in which some or all the light source units 101 located in the same shape as the display image of the plurality of light source units 101 included in the irradiation unit 10 radiate light.

Hereinafter, an operation in which the irradiation unit 10 displays an image having a size larger than a predetermined size, such as displaying a linear image or displaying a plurality of figures, is referred to as a linear image display operation. Hereinafter, the operation in which the irradiation unit 10 displays an image of a single point smaller than a predetermined size by the irradiation unit 10 is referred to as a single point display operation.
When the irradiation unit 10 executes the linear image display operation, a wider range of the inspection target 9 is irradiated at one time than when the irradiation unit 10 executes the single point display operation.

The linear image display operation is, for example, an operation in which some or all of the light source units 101 arranged in a line among the plurality of light source units 101 included in the irradiation unit 10 emit light.
The single point display operation is, for example, an operation in which some or all of the light source units 101 located within a predetermined circumference of the plurality of light source units 101 included in the irradiation unit 10 emit light.

The irradiation unit 10 may be any surface emitting light source. The irradiation unit 10 may be, for example, a rear projection display using a liquid crystal display, a digital mirror device, or the like. The irradiation unit 10 may be, for example, an organic EL (electro-luminescence) display. The irradiation unit 10 may be, for example, a light source in which one linear polarizing plate is arranged so as to cover the entire front surface of the organic EL display.

The polarizing filter 20 is a polarizing filter whose filter direction can be changed, and extracts and emits polarized light whose polarization direction is the filter direction from incident light. The filter direction is the polarization direction of the light transmitted through the polarizing filter 20. The polarizing filter 20 is, for example, a rotatable polarizing filter that changes the filter direction by rotating.
For example, the polarizing filter 20 emits polarized light whose polarization direction coincides with the filter direction when the reflected light or scattered light of 1 L of light incident on the inspection target 9 is incident. The rotation of the polarizing filter 20 changes the filter direction. In the inspection system 1, the polarization direction dependence of the light incident on the polarizing filter 20 is measured by changing the filter direction. The polarizing filter 20 may change the filter direction by any mechanism as long as the filter direction can be changed, and it is not always necessary to change the filter direction by rotating the polarizing filter 20.

The image pickup apparatus 40 includes a plurality of light receiving elements 401 arranged two-dimensionally. The image pickup apparatus 40 receives a plurality of secondary lights, which are the reflected light or scattered light of 1 L of light transmitted through the polarizing filter 20 and incident on the inspection target 9, and which is the reflected light or scattered light transmitted through the polarizing filter 20. Light is received by the element 401.

The image pickup apparatus 40 generates image set information for each of the plurality of filter directions based on the received secondary light. The image set information is a set of image data of two-dimensional images for each of a plurality of predetermined light wavelengths. Each of the two-dimensional images for each light wavelength is an image showing the intensity of the wavelength component of the secondary light received by each light receiving element 401 for each light wavelength. The light wavelength is the wavelength of light.

As described above, in the inspection system 1, the secondary light received by the image pickup apparatus 40 is polarized light that has passed through the polarizing filter 20 among the secondary light of 1 L of light incident on the inspection target 9. In the inspection system 1, the image pickup apparatus 40 generates image set information based on the received polarized light.

The image pickup apparatus 40 is, for example, a hyperspectral camera. The image set information is, for example, a data cube.

The polarizing filter 20 does not necessarily have to be located outside the housing of the image pickup apparatus 40. The polarizing filter 20 may be located at any position as long as it is located between the emission position of the secondary light of 1 L of light and the light receiving element 401. For example, the polarizing filter 20 may be located inside the housing of the imaging device 40. The polarizing filter 20 and the image pickup apparatus 40 may be integrally configured. Hereinafter, the functional unit in which the polarizing filter 20 and the imaging device 40 are integrally configured is referred to as an imaging unit 41.

The image pickup unit 41 may have, for example, a rotatable linear polarizing plate arranged on the front surface of the lens of a digital camera. The image pickup unit 41 may be more preferably, for example, the polarized camera described in References 1 to 3 described below.
Reference 1: Japanese Patent Application Laid-Open No. 2017-017563 Reference 2: Japanese Patent Application Laid-Open No. 2014-57231 Reference Reference 3: Japanese Patent Application Laid-Open No. 2012-80065

The imaging unit 41 may be, for example, VP-PHX050S-P manufactured by LUCID, Nano-M2450-Polarized manufactured by Teledyne DALSA, or Blackfly S Polarized 5.0 MP USB3 Vision manufactured by FLIR.

The inspection device 50 includes a CPU (Central Processing Unit) 501, a memory 502, an auxiliary storage device 503, and the like connected by a bus, and executes a program. The inspection device 50 functions as a device including an input unit 504 and an output unit 505 by executing a program.
The CPU 501 functions as a detection unit 51 by executing a program stored in the memory 502 or the auxiliary storage device 503. The detection unit 51 detects a defective portion of the appearance of the inspection target 9 based on the image set information. To detect a defective part in the appearance of the inspection target 9, a process of determining whether or not there is a defective part in the appearance of the inspection target 9, and a process of identifying the position of the defective part when it is determined that there is a defective part. Is executed by the detection unit 51.

The inspection system control unit 60 includes a CPU, a memory, an auxiliary storage device, and the like, and executes a program. The inspection system control unit 60 controls the operation of each functional unit included in the inspection system 1 by executing a program. The inspection system control unit 60 causes the image pickup device 40 to generate image set information for each of a plurality of filter directions by, for example, operating the polarizing filter 20 and the image pickup device 40 in synchronization with each other when the irradiation unit 10a is irradiated.

The light source unit 101, the inspection target 9, the polarizing filter 20, and the imaging device 40 use the polarizing filter 20 on the surface (irradiated surface) of the inspection target 9 irradiated by the radiation of light by the light source unit 101. It may be located at any position as long as it is located at a position where the image pickup apparatus 40 can take an image.

The positional relationship of the plurality of light source units 101 is such that the area of the irradiated surface on the inspection target 9 is maximized when all of the light source units 101 emit light among the positional relationships of the plurality of light source units 101. Therefore, it is desirable that the image pickup apparatus 40 is located in a direction in which the light emitted on the inspection target 9 is specularly reflected. In such a case, the light intensity of the reflected light incident on the image pickup apparatus 40 becomes stronger, and the time required for the inspection becomes shorter. Hereinafter, such a positional relationship between the light source unit 101, the inspection target 9, the polarizing filter 20, and the image pickup apparatus 40 will be referred to as a first ideal positional relationship.
It is more desirable that the image pickup apparatus 40 is located at a distance at which the entire irradiated surface can be photographed. In such a case, the area that can be inspected in one shooting is maximized, and the time required for the inspection is shortened. Hereinafter, among the first ideal positional relationships, the positional relationship in which the imaging device 40 is located at a distance at which the entire irradiated surface can be photographed is referred to as a second ideal positional relationship.

Here, the outline of the principle relating to the inspection system 1 will be described with reference to FIG.
FIG. 2 is an explanatory diagram illustrating an outline of the principle according to the inspection system 1 of the embodiment. In FIG. 2, the irradiation unit 10a is a liquid crystal display. In FIG. 2, the inspection target 9 is a transparent or translucent object.

The directly reflected light 1L by the inspection target 9 of the light illuminated by one bright spot on the irradiation unit 10a includes mirror-reflected light, diffuse-reflected light, and light due to subsurface scattering (hereinafter referred to as "subsurface scattered light"). .) And are included. The specularly reflected light is polarized, but the diffusely reflected light and the subsurface scattered light are not polarized due to multiple scattering.

In general, the transmitted intensity of polarized light changes with the rotation of the polarizing plate. On the other hand, the transmission intensity of unpolarized light does not change due to the rotation of the polarizing plate. Therefore, in the inspection system 1, the light intensity that fluctuates according to the rotation of the polarizing plate among the light intensity of the directly reflected light 1L is the light intensity derived from the specularly reflected light. Therefore, if the light intensity that fluctuates according to the rotation of the polarizing plate and the light intensity that does not fluctuate are measured, the light intensity of the specular reflected light and the light intensity of the diffuse reflected light and the subsurface scattered light can be separated. ..

In the inspection system 1, the polarizing filter 20 is a polarizing plate. In the inspection system 1, if the light intensity in four or more different filter directions is measured by rotating the polarizing filter 20, the light intensity of the specular reflected light and the light intensity of the diffuse reflected light and the subsurface scattered light are measured. Is separable from. Hereinafter, the light intensity of the specularly reflected light is referred to as a specularly reflected light component. Hereinafter, the light intensity of the diffuse reflected light and the subsurface scattered light is referred to as a non-polarized component. The specular reflected light, the diffuse reflected light, and the subsurface scattered light are all secondary lights. The directly reflected light 1L is an example of secondary light.

Specifically, in the inspection system 1, the specularly reflected light component and the non-polarized light component are separated by executing the following processing.
In the inspection system 1, first, image set information for each of a plurality of filter directions is acquired. Hereinafter, the image set information for each of a plurality of filter directions is referred to as a filter transmission result.

In the inspection system 1, the inspection apparatus 50 then calculates the maximum value Imax and the minimum value Imin of the light intensity based on the filter transmission result by, for example, the method described in Reference 4 described below. Next, the inspection device 50 separates the specularly reflected light component and the non-polarized light component based on the maximum value Imax and the minimum value Imin.
Reference 4: Lawrence B Wolff and Terrance E. Boult "Constraining object features using a polarization reflectance model", IEEE Trans. PAMI, 1991

The light intensity Is of the specularly reflected light calculated by the inspection device 50 is represented by the following formula (1).

Next, the inspection device 50 calculates the value of the surface roughness parameter σ, which is a value representing the surface roughness, which is a parameter representing the surface roughness. The surface roughness parameter σ represents the surface roughness depending on the magnitude of the value. The surface roughness parameter σ may indicate that the surface is rough as the value is large, or that the surface is rough as the value is small.
The surface roughness parameter σ is, for example, a value representing the surface roughness in a predetermined reflection model. The predetermined reflection model is, for example, a simplified transformer sparrow model, a ward model considering the anisotropy of the surface roughness parameter of the surface to be inspected, a phone model, a lafortune model, and the like. The value representing the surface roughness in the predetermined reflection model may be, for example, a parameter representing the surface roughness in the simplified trans-sparrow model.
Hereinafter, for the sake of simplicity, it is assumed that the larger the value of the surface roughness parameter σ, the rougher the surface.

The surface roughness parameter σ is expressed by, for example, the following equation (2). The surface roughness parameter σ represented by the equation (2) is a parameter representing the surface roughness in the simplified trans-sparrow model.

The trans-sparrow model is described, for example, in reference 5 below.
Reference 5: KE Torrance and EM Sparrow, “Theory for off-specular reflection from roughened surfaces”, Journal of the Optical Society of America, Vol.57, No.9, pp.1105-1114, 1967

In equation (2), N is a vector representing the direction of the object normal in the simplified trans-sparrow model, as shown in FIG. In equation (2), V is a vector representing the viewpoint direction in the simplified trans-sparrow model, as shown in FIG. In formula (2), Ks represents the mirror reflectance of the inspection target 9 in the simplified trans-sparrow model. In the formula (2), as shown in FIG. 2, β is the bisection direction H of the angle formed by the direction of the irradiation unit 10 and the direction of the imaging device 40 in the simplified trans-sparrow model, and the inspection target 9 It is the angle formed by the direction N of the object normal of. The eggplant angle β is a value obtained by the method described on page 22 of Non-Patent Document 1.
The mirror reflectance Ks of the inspection target 9 is calculated based on the mirror reflectance of surface-smooth glass having a refractive index of 1.567, for example, by the method described in Reference 6 or Reference 7 described below.
Reference 6: "Japanese Industrial Standards Mirror Gloss-Measuring Method" JIS Z 8741: 1997
Reference 7: "Japanese Industrial Standard Paint General Test Method-Part 4: Visual Characteristics of Coating Film-Section 7: Mirror Gloss" JIS K 5600-4-7: 1999

The inspection device 50 detects a defective portion of the appearance of the inspection target 9 based on the surface roughness parameter σ.

FIG. 3 is a diagram showing an example of the functional configuration of the detection unit 51 in the embodiment. The detection unit 51 executes the detection process. The detection unit 51 detects a defective portion of the appearance of the inspection target 9 by executing the detection process.

The detection unit 51 includes an image generation unit 511, a separation unit 512, a surface roughness calculation unit 513, a defective pixel determination unit 514, and a defective pixel presence / absence determination unit 515.

The image generation unit 511 generates an image of the inspection target 9 (hereinafter referred to as “inspection target image”) based on the image set information. The image to be inspected is an image in which the brightness of the pixel corresponds to the intensity of the secondary light received by the light receiving element 401. The image to be inspected is an image in which the brightness of each pixel corresponds to the intensity of the secondary light received by one light receiving element 401. The image to be inspected is an image generated based on the intensity of the secondary light received by at least two light receiving elements 401. For example, any two pixels of the image to be inspected are images having brightness corresponding to the intensity of the secondary light received by different light receiving elements 401.

The separation unit 512 separates the specularly reflected light component and the non-polarized light component in each pixel of the image to be inspected based on the image set information for each of the plurality of light emitting light sources. The light emitting light source is a light source unit 101 that emits light. Separating the specularly reflected light component and the non-polarized light component is a process of acquiring the light intensity of the specularly reflected light.

The separation unit 512 acquires position wavelength spectrum information by separating the specularly reflected light component and the non-polarized light component in each pixel. The position wavelength spectrum information is information indicating the specularly reflected light component of each pixel of the image to be inspected (that is, the light intensity Is of the specularly reflected light in each pixel).

The surface roughness calculation unit 513 calculates the value of the surface roughness parameter σ of the inspection target 9 for each pixel of the inspection target image based on the position wavelength spectrum information.

The surface roughness calculation unit 513 calculates, for example, the light intensity Is represented by the formula (1) and the value of the surface roughness parameter σ represented by the formula (2) based on the position wavelength spectrum information. Then, the value of the surface roughness parameter σ is acquired for each pixel of the image to be inspected.

The defective pixel determination unit 514 determines whether or not the pixel is a defective pixel for each pixel of the inspection target image based on the surface roughness parameter σ of each pixel acquired by the surface roughness calculation unit 513. The defective pixel is determined by using a predetermined roughness reference value for each pixel. Hereinafter, for the sake of simplicity, it is assumed that the defective pixel is a pixel whose surface roughness parameter σ is equal to or greater than the roughness reference value, but it can be similarly determined even if it is a pixel having a roughness reference value or less.
Hereinafter, the pixel determined by the defective pixel determination unit 514 to be not a defective pixel is referred to as a normal pixel. That is, a normal pixel is a pixel whose surface roughness parameter σ is less than the roughness reference value.

The defective pixel determination unit 514 outputs information indicating the determination result. The information indicating the determination result may be any type of information as long as it is information indicating defective pixels and normal pixels. For example, the determination result may be an image in which the pixels corresponding to the defective pixels of the inspection target image are highlighted.
In such a case, if the auxiliary storage device 503 stores predetermined information (hereinafter referred to as “correspondence information”) for associating the pixels with the light receiving element 401, the detection unit 51 does not necessarily have to be the image generation unit 511. There is no need to prepare. The correspondence information does not necessarily have to be stored in the auxiliary storage device 503, and may be input via the input unit 504 from an external device (not shown) that stores the correspondence information each time the inspection device 50 operates. Good.

The defective pixel presence / absence determination unit 515 determines the presence / absence of defective pixels based on the determination result of the defective pixel determination unit 514. The defect pixel presence / absence determination unit 515 determines that there are defective pixels when there are defective pixels. The defect pixel presence / absence determination unit 515 determines that there are no defective pixels when there are no defective pixels.

FIG. 4 is a flowchart showing an example of processing executed by the inspection system 1 of the embodiment.
Under the control of the inspection system control unit 60, the irradiation unit 10 operates and irradiates light (step S101).
The irradiation unit 10, the polarizing filter 20, and the imaging device 40 operate under the control of the inspection system control unit 60, and the imaging device 40 generates image set information for each of the plurality of filter directions of the polarizing filter 20 (step S102). ..

The separation unit 512 separates the specularly reflected light component and the non-polarized light component in each pixel of the image to be inspected based on the image set information in each filter direction generated by the image pickup apparatus 40. The separation unit 512 acquires position wavelength spectrum information by separating the specularly reflected light component and the non-polarized light component in each pixel (step S103).
The surface roughness calculation unit 513 calculates the value of the surface roughness parameter σ of the inspection target 9 for each pixel of the inspection target image based on the position wavelength spectrum information (step S104).

The defective pixel determination unit 514 determines whether or not the pixel is a defective pixel for each pixel of the inspection target image based on the surface roughness parameter σ of each pixel acquired by the surface roughness calculation unit 513 (step S105).
Specifically, the defective pixel determination unit 514 determines for each pixel whether or not the surface roughness parameter σ is equal to or greater than the roughness reference value of each pixel. The defective pixel determination unit 514 determines that a pixel whose surface roughness parameter σ is equal to or greater than the roughness reference value of each pixel is a defective pixel. The defective pixel determination unit 514 determines that a pixel whose surface roughness parameter σ is less than the roughness reference value of each pixel is a normal pixel.

The defective pixel presence / absence determination unit 515 determines whether or not there are defective pixels based on the determination result of the defective pixel determination unit 514 (step S106). Specifically, the defective pixel presence / absence determination unit 515 determines that there are defective pixels when the determination result of the defective pixel determination unit 514 includes defective pixels. The defective pixel presence / absence determination unit 515 determines that there are no defective pixels when there are no defective pixels in the determination result of the defective pixel determination unit 514.

FIG. 5 is a diagram showing an example of the experimental result of the inspection target image in the embodiment.
In the experimental system giving the experimental results of FIG. 5, the irradiation unit 10 was a liquid crystal display. In the experimental system giving the experimental results of FIG. 5, the inspection target 9 was frosted glass. In the experimental system giving the experimental results of FIG. 5, the imaging unit 41 was a VP-PHX050SP manufactured by LUCID. In the experimental system giving the experimental results of FIG. 5, the positional relationship between the irradiation unit 10, the inspection target 9, and the imaging unit 41 was the second ideal positional relationship.

The experimental result of FIG. 5 is an example of a result in which a liquid crystal display displays white lines having a width of 5 pixels on the display in order from the top, and each time the white lines are displayed on frosted glass, the state of the white lines is photographed by the imaging device 40. Is.

The inspection system 1 configured in this way includes an irradiation unit 10 including a plurality of light source units 101, and irradiates the inspection target 9 with the light emitted by the light source unit 101. The inspection system 1 includes a polarizing filter 20 whose filter direction can be changed, and receives the secondary light of the light irradiated to the inspection target 9 through the polarizing filter 20. The inspection system 1 separates the specularly reflected light component and the non-polarized light component based on the result of receiving the light. Therefore, the inspection system 1 configured in this way can reduce the influence of the non-polarizing component and detect the defect in the appearance of the inspection object, thus improving the accuracy of calculating the parameter representing the surface roughness. be able to. Further, the inspection system 1 configured in this way can reduce the influence of the non-polarizing component and detect the defect in the appearance of the inspection object, so that the defect detection accuracy can be improved.

Further, in the inspection system 1 configured in this way, since the irradiation unit 10 includes a plurality of light source units 101 having different positions, the incident angle of the light irradiating the inspection target 9 can be changed. Therefore, it is possible to reduce the frequency with which the image pickup apparatus 40 cannot receive the specularly reflected light depending on the inspection target.

Further, in the inspection system 1 configured in this way, since the irradiation unit 10 includes a plurality of light source units 101 having different positions, the incident angle of the light irradiated to the inspection target 9 can be changed without moving the irradiation unit 10. It is not necessary to move the irradiation unit 10. Therefore, the inspection system 1 configured in this way can improve the accuracy of calculating the parameter representing the surface roughness while reducing the energy required for the movement of the irradiation unit 10.

(Modification example)
FIG. 6 is a diagram showing an example of the functional configuration of the inspection system 1a in the modified example.
The inspection system 1a differs from the inspection system 1 in that it includes an imaging unit drive unit 70.
Hereinafter, those having the same functions as the inspection system 1 will be designated by the same reference numerals as those in FIG. 1, and the description thereof will be omitted.

The image pickup unit drive unit 70 moves the image pickup device 40 under the control of the inspection system control unit 60. The image pickup unit drive unit 70 may be any type as long as the image pickup device 40 can be moved. The image pickup unit drive unit 70 may be, for example, a device that moves the image pickup device 40 by the rotation of the stepping motor.

When the image pickup device 40 receives secondary light at a plurality of positions by being moved by the image pickup unit drive unit 70, the image set information in the inspection system 1a is the wavelength spectrum of each light receiving element 401 at each position of the image pickup device 40. It may be information indicating.

In the inspection system 1a, the imaging device 40 does not necessarily have to include a plurality of light receiving elements 401, and may include only one light receiving element 401.

The detection unit 51 of the inspection systems 1 and 1a may include a restoration unit 516.
FIG. 7 is a diagram showing an example of the restoration unit 516 in the modified example.
The restoration unit 516 acquires the wavelength spectrum of each light receiving element 401 when only one light source unit 101 emits light by the method of multiplexing measurement. The method of multiplexing measurement may be, for example, the method disclosed in Reference 8 described below.
Reference 8: Yoav Y. Schechner, Shree K. Nayar and Peter N. Belhumeur "A Theory of Multiplexed Illumination", Proceedings of the Ninth IEEE International Conference on Computer Vision (ICCV'03), 0-7695-1950-4 / 03, 2003

The surface roughness parameter σ does not necessarily have to be a value represented by the equations (1) and (2). The surface roughness parameter σ may be, for example, a value representing a change in the specular reflection component with respect to a change in the angle β formed.

Hereinafter, the change in the specular reflection component will be described with reference to FIG. 8 by taking the experimental system described with reference to FIG. 5 as a specific example. For example, when the liquid crystal display displays white lines in order on the display with the display state of the liquid crystal display such that the angle β formed by a predetermined pixel is 3.5 degrees as the initial state, the angle β formed by the pixel is the initial state. It changes from the angle β formed in the state. Further, when the liquid crystal display sequentially displays white lines on the display, the specular reflection component of the secondary light incident on the image pickup apparatus 40 also changes.

FIG. 8 is a diagram showing an example of experimental results showing the correspondence between the change in the angle β formed in the modified example and the change in the specular reflection component.
The horizontal axis of FIG. 8 represents the angle β formed. The vertical axis of FIG. 8 represents the light intensity. FIG. 8 shows the relationship between the angle β formed in a predetermined pixel and the light intensity of the specular reflection component. FIG. 8 shows the relationship between the angle β formed in a predetermined pixel and the light intensity of a plurality of reflection components. FIG. 8 shows that the light intensity of the specular reflection component changes according to the change in the angle β formed.
Further, FIG. 8 shows that β has a peak that reaches its maximum near 3.5 degrees. Hereinafter, it is referred to as an angle spectrum forming information indicating the light intensity with respect to the angle β (for example, the graph shown in FIG. 8). The angle spectrum formed has a peak that maximizes the light intensity as shown in FIG.

By the way, the smaller the surface roughness of the inspection target 9, and the smaller the influence of subsurface scattering, the more the light intensity of the reflected light is dominated by the directly reflected light, and the angle spectrum of the light incident on the image pickup apparatus 40. The shape of the peak becomes sharp and narrow. On the contrary, the larger the surface roughness or the stronger the influence of subsurface scattering, the more isotropic the intensity shape of the reflected light. Therefore, the shape of the peak of the angle spectrum formed becomes wide.
As described above, the shape of the peak of the angle spectrum formed is a shape corresponding to the roughness of the surface. Therefore, the surface roughness parameter σ may be a value representing the shape of the peak of the angle spectrum formed.

When the value representing the shape of the peak of the angle spectrum is used as the surface roughness parameter σ, the information indicating the relationship between the value obtained by subtracting the background value in the experimental result from the light intensity of the measurement result and the angle formed. It is desirable to have an angular spectrum that forms.

For example, in FIG. 8, the shape of the angle spectrum formed by subtracting the value of the background in which the angle β forming is 2 degrees or less and the angle forming 4 degrees or more from the intensity of the peak is the peak intensity. It is desirable that the surface roughness parameter σ represents it.

Hereinafter, the inspection system 1 in the case of expressing the shape of the angular spectrum formed by the surface roughness parameter σ is referred to as an inspection system 1b.
The value representing the shape of the forming angle spectrum may be, for example, the half width of the peak of the forming angle spectrum or the intensity of the peak.
When using the half width of the peak, it is not necessary to separately obtain the mirror reflectance.

FIG. 9 is a diagram showing an example of the functional configuration of the inspection system 1b in the modified example. The inspection system 1b is different from the inspection system 1 in that the image pickup device 40b is provided in place of the image pickup device 40 and the inspection device 50b is provided in place of the inspection device 50. Hereinafter, those having the same functions as the inspection system 1 will be designated by the same reference numerals as those in FIG. 1, and the description thereof will be omitted.

The image pickup apparatus 40b is different from the image pickup apparatus 40 in that it generates a large image set information instead of the image set information. The large image set information is the image set information for each display image of the irradiation unit 10 and for each filter direction.

The inspection device 50b is different from the inspection device 50 in that the detection unit 51b is provided in place of the detection unit 51.
FIG. 10 is a diagram showing an example of the functional configuration of the detection unit 51b in the modified example. The detection unit 51b includes an image generation unit 511b instead of the image generation unit 511, a separation unit 512b instead of the separation unit 512, and a surface roughness calculation unit 513b instead of the surface roughness calculation unit 513. It is different from the detection unit 51 in that it is provided with. Hereinafter, those having the same function as the detection unit 51 are designated by the same reference numerals as those in FIG. 3, and the description thereof will be omitted.

The image generation unit 511b differs from the image generation unit 511 in that an inspection target image is generated for each display image based on the large image set information instead of the image set information.

The separation unit 512b acquires position wavelength spectrum information for each display image by separating the specular reflected light component and the non-polarized light component in each pixel based on the large image set information instead of the image set information. It is different from the separation unit 512. Hereinafter, the position wavelength spectrum information for each display image is referred to as a large position wavelength spectrum information.

The surface roughness calculation unit 513b calculates the value of the surface roughness parameter σ of the inspection target 9 for each pixel of the inspection target image based on the large position wavelength spectrum information instead of the position wavelength spectrum information. It is different from the roughness calculation unit 513. The surface roughness parameter σ calculated by the surface roughness calculation unit 513b is a value representing the shape of the angle spectrum formed by the surface roughness calculation unit 513b, and has a definition different from the surface roughness parameter σ calculated by the surface roughness calculation unit 513. Is.

FIG. 11 is a flowchart showing an example of processing executed by the inspection system 1b of the modified example. Hereinafter, the same processing as that of the flowchart shown in FIG. 4 will be described by adding the same reference numerals.

The irradiation unit 10 operates under the control of the inspection system control unit 60, and by displaying a plurality of display images, the inspection target 9 is irradiated with light corresponding to each display image (step S201).
Under the control of the inspection system control unit 60, the irradiation unit 10, the polarizing filter 20, and the image pickup device 40b operate, and the image pickup device 40b generates a large image set information (step S202).

The separation unit 512b separates the specularly reflected light component and the non-polarized light component in each pixel of the image to be inspected based on the large image set information generated by the image pickup apparatus 40b. The separation unit 512b acquires large-position wavelength spectrum information by separating the specularly reflected light component and the non-polarized light component in each pixel (step S203).
The surface roughness calculation unit 513b calculates the value of the surface roughness parameter σ of the inspection target 9 for each pixel of the inspection target image based on the large position wavelength spectrum information (step S204).

The inspection system 1b of the modified example configured as described above includes an irradiation unit 10 including a plurality of light source units 101, and irradiates the inspection target 9 with the light emitted by the light source unit 101. The inspection system 1b includes a polarizing filter 20 whose filter direction can be changed, and receives the secondary light of the light irradiated to the inspection target 9 through the polarizing filter 20. The inspection system 1b separates the specularly reflected light component and the non-polarized light component based on the result of receiving the light. Therefore, the inspection system 1b configured in this way can reduce the influence of the non-polarizing component and detect the defect in the appearance of the inspection object, thus improving the accuracy of calculating the parameter representing the surface roughness. be able to. Further, the inspection system 1b configured in this way can reduce the influence of the non-polarizing component and detect the defect in the appearance of the inspection object, so that the defect detection accuracy can be improved.

The surface roughness calculation unit 513b may execute an operation for reducing the influence of the background of the angle spectrum formed. For example, when the polarizing filter 20 extracts linearly polarized light, by acquiring only the intensity of a predetermined polarizing component based on the equation (1), as shown in the graph of “plurality of reflection components” in FIG. The influence of the ground is significantly reduced.

The imaging device 40 is an example of an imaging unit.

All or part of the functions of the inspection system 1 and the inspection system 1a are realized by using hardware such as ASIC (Application Specific Integrated Circuit), PLD (Programmable Logic Device), and FPGA (Field Programmable Gate Array). You may. The program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a flexible disk, a magneto-optical disk, a portable medium such as a ROM or a CD-ROM, or a storage device such as a hard disk built in a computer system. The program may be transmitted over a telecommunication line.

The inspection device 50 may be mounted by using a plurality of information processing devices that are communicably connected via a network. In this case, each functional unit included in the inspection device 50 may be distributed and mounted in a plurality of information processing devices. For example, the image generation unit 511, the separation unit 512, the surface roughness calculation unit 513, the defective pixel determination unit 514, the defective pixel presence / absence determination unit 515, and the restoration unit 516 are mounted on different information processing devices. May be good.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and the design and the like within a range not deviating from the gist of the present invention are also included.

1, 1a, 1b ... Inspection system, 10 ... Irradiation unit, 20 ... Polarization filter, 40, 40b ... Imaging device, 50, 50b ... Inspection device, 60 ... Inspection system control unit, 70 ... Imaging unit drive unit, 401 ... Light receiving Element, 501 ... CPU (Central Processing Unit), 502 ... Memory, 503 ... Auxiliary storage device, 504 ... Input unit, 505 ... Output unit, 51, 51b ... Detection unit, 511, 511b ... Image generation unit, 512, 512b ... Separation unit, 513, 513b ... Surface roughness calculation unit, 514 ... Defective pixel determination unit, 515 ... Defective pixel presence / absence determination unit, 516 ... Restoration unit

Claims (8)

An irradiation unit including a plurality of light source units that are located in a predetermined plane and emit linearly polarized light in a predetermined polarization direction.
The light emitted by the irradiation unit is mirror-reflected, diffusely reflected, or subsurfacely scattered by the inspection target, and is polarized light extracted by a polarizing filter, and a plurality of polarized light among the secondary light. An imaging unit that receives polarized light in a direction and generates image set information, which is a set of image data of two-dimensional images for each of a plurality of predetermined light wavelengths, based on the intensity of the received polarized light for each polarized direction. When,
A separation unit that acquires a wavelength spectrum of specularly reflected light based on the plurality of image set information generated by the imaging unit, and a separation unit.
A surface roughness calculation unit that calculates a surface roughness parameter representing the surface roughness of the inspection target based on the wavelength spectrum acquired by the separation unit.
Inspection device equipped with. The surface roughness calculation unit calculates the surface roughness parameter based on the wavelength spectrum for each polarization direction.
The inspection device according to claim 1. The surface roughness parameter is a surface roughness parameter in a predetermined reflection model.
The inspection device according to claim 2. The predetermined reflection model is a simplified transformer sparrow model.
The inspection device according to claim 3. A restoration unit that acquires the wavelength spectrum of the secondary light when only one of the plurality of light source units emits light by the method of multiplexing measurement.
Further prepare
The inspection device according to any one of claims 1 to 4. The surface is defined as an angle β forming an angle formed by an bisector of an angle formed by the direction of the irradiation portion and the direction of the imaging portion in the simplified trans-sparrow model and the direction of the object normal to be inspected. The roughness parameter is a value representing the shape of the peak of the intensity in the angle spectrum, which is information indicating the dependence of the intensity received by the imaging unit on the angle β formed.
The inspection device according to claim 2. An irradiation unit including a plurality of light source units that are located in a predetermined plane and emit linearly polarized light in a predetermined polarization direction.
A polarizing filter that extracts polarized light in a predetermined polarization direction,
The light irradiated by the irradiation unit is mirror-reflected, diffusely reflected, or subsurfacely scattered by the inspection target, and is polarized light extracted by the polarizing filter, with the light being scattered under the surface as the secondary light, and is a plurality of the secondary lights. Imaging that receives polarized light in the polarization direction and generates image set information that is a set of image data of two-dimensional images for each of a plurality of predetermined light wavelengths based on the intensity of the received polarized light for each polarization direction. Department and
A separation unit that acquires a wavelength spectrum of specularly reflected light based on the plurality of image set information generated by the imaging unit, and a separation unit.
A surface roughness calculation unit that calculates a surface roughness parameter representing the surface roughness of the inspection target based on the wavelength spectrum acquired by the separation unit.
Inspection system equipped with. An irradiation step in which an irradiation unit including a plurality of light source units located in a predetermined plane and emitting linearly polarized light in a predetermined polarization direction irradiates the light.
The light emitted by the irradiation unit is mirror-reflected, diffusely reflected, or subsurfacely scattered by the inspection target, and is polarized light extracted by a polarizing filter, and a plurality of polarized light among the secondary light. An imaging step that receives polarized light in a direction and generates image set information that is a set of image data of two-dimensional images for each of a plurality of predetermined light wavelengths based on the intensity of the received polarized light for each polarized light direction. When,
A separation step for acquiring a wavelength spectrum of specularly reflected light based on the plurality of image set information generated in the imaging step, and a separation step.
A surface roughness calculation step for calculating a surface roughness parameter representing the surface roughness of the inspection target based on the wavelength spectrum acquired in the separation step, and a surface roughness calculation step.
Inspection method with.