JP2021135200A - Image inspection device and image inspection method - Google Patents

Abstract

To provide an image inspection device that enables a device used to inspect an object by photographing one image, to be more freely arranged.SOLUTION: An image inspection device (1, 2) comprises: a projection part (10) which irradiates an object with light having a striped pattern with a uniform pitch to project an image of the striped pattern on the object; a photography part (20) which photographs the image of the striped pattern projected on the object; a shape information calculation part (320) which calculates shape information on the object based upon image features extracted from the striped pattern included in the image; and an inspection part (330) which inspects the object based upon the shape information, the striped pattern included in the image having an irregular pitch.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image inspection device and an image inspection method.

Conventionally, there is known a technique of irradiating an object with light from a light source, projecting an image, and measuring the shape of the object using an image obtained by photographing the projected image. For example, Patent Document 1 describes a technique of projecting a lattice image on an object surface, obtaining the phase of a specific frequency component in the lattice image, and measuring the position of the object surface using the phase.

International Publication No. 2016/001985

A method using the phase shift method is also known, but the phase shift method requires taking a plurality of images. On the other hand, in the technique described in Patent Document 1, since the shape of an object can be measured with one image, the time required for measurement is shortened as compared with the phase shift method in which a plurality of images need to be taken. NS.

However, the technique described in Patent Document 1 is based on the premise that the pitch of the lattice image projected on the surface of the object is uniform. Therefore, in the technique described in Patent Document 1, the degree of freedom in arranging a device used for measurement, such as a light source for projecting an image or a camera for capturing an image, is low.

Therefore, an object of the present invention is to provide an image inspection device and an image inspection method that enable more freely arranging devices used for inspecting an object by taking a single image.

The image inspection apparatus according to one aspect of the present invention has a projection unit that irradiates an object with a striped pattern of light having a uniform pitch and projects an image of the striped pattern onto the object, and a striped pattern projected onto the object. An imaging unit that captures an image of The striped pattern included in the image has a non-uniform pitch.

According to this aspect, the object can be inspected by taking one image. Further, the arrangement of the projection unit and the photographing unit can be appropriately changed while maintaining the non-uniform pitch of the fringe pattern projected on the object. Therefore, the arrangement of the device used for inspecting the object becomes more free than when the image of the striped pattern having a uniform pitch is projected on the object.

In the above aspect, the inspection unit may detect the position of the abnormality in the object.

According to this aspect, it is possible to inspect the object in more detail.

In the above aspect, the shape information calculation unit may generate the first shape image by calculating the shape information corresponding to each of the plurality of positions in the striped pattern image and mapping the calculated shape information in two dimensions. good.

According to this aspect, more appropriate inspection can be performed by using the shape information.

In the above aspect, the inspection unit may detect an abnormality of the object based on the second shape image in which the shape of the specific article is represented in two dimensions and the first shape image.

According to this aspect, the object can be inspected more appropriately.

In the above aspect, a correction unit for correcting the positional deviation of the first shape image and the second shape image may be further provided.

According to this aspect, even when there is a positional deviation between the first shape image and the second shape image before correction, the inspection unit can appropriately inspect the object.

In the above aspect, the inspection unit binarizes the shape information included in the first shape image with a threshold value, and determines the object based on the area corresponding to any of the binary values in the first shape image. You may inspect.

According to this aspect, the inspection unit can inspect the object more appropriately.

In the above aspect, the projection unit is a flat plate-shaped illumination that irradiates diffused light, and the object may be irradiated with light having a striped pattern.

According to this aspect, it becomes possible to irradiate the light of the striped pattern more easily.

In the above aspect, the projection unit irradiates the object with light of a plurality of stripe patterns having different stripe directions, the photographing unit captures each image of the plurality of stripe patterns, and the shape information calculation unit has a plurality of images. Shape information may be calculated for each image of the fringe pattern of the above, and the shape information of the plurality of calculated fringe patterns may be combined, and the inspection unit may inspect the object based on the combined shape information.

According to this aspect, since the inspection unit can use the shape information based on the images of the plurality of striped patterns, it is possible to inspect a more appropriate object.

An image inspection method according to another aspect of the present invention is to irradiate an object with light of a striped pattern having a uniform pitch and project an image of the striped pattern onto the object, and a striped pattern projected onto the object. The image includes taking an image of the object, calculating the shape information of the object based on the image feature extracted from the stripe pattern in the image, and inspecting the object based on the shape information. The included striped pattern has a non-uniform pitch.

According to this aspect, the object can be inspected by taking one image. Further, the arrangement of the projection unit and the photographing unit can be appropriately changed while maintaining the non-uniform pitch of the fringe pattern projected on the object. Therefore, the arrangement of the device used for inspecting the object becomes more free than when the image of the striped pattern having a uniform pitch is projected on the object.

According to the present invention, it is possible to provide an image inspection device and an image inspection method that enable more freely arranging devices used for inspecting an object by taking one image.

It is a figure which shows the structure of the image inspection apparatus which concerns on one Embodiment of this invention. It is a figure which shows an example of the image of the stripe pattern which the photographing part takes. It is a functional block diagram which shows an example of the structure of a processing part. It is a functional block diagram which shows the structure of the shape information calculation part which concerns on one Embodiment of this invention. It is a figure which shows the pitch distribution image generated by the pitch estimation part performing the pitch estimation process on the image shown in FIG. It is a figure which shows the 1st phase image which shows the phase distribution, which was generated by the phase calculation part using the pitch distribution image shown in FIG. It is a figure which shows the 2nd phase image generated by connecting the phase of the 1st phase image shown in FIG. 6 by a phase connection part. It is a figure for demonstrating the method which the angle calculation part calculates shape information. It is a figure which shows the shape image generated by mapping the angle Ψ calculated by the image generation part in two dimensions. It is a figure which shows the binary image generated by the inspection part binarizing the angle Ψ included in the shape image shown in FIG. It is a figure which shows the physical structure of the processing part which concerns on this embodiment. It is a flowchart which shows the process of the image inspection apparatus which concerns on this embodiment. It is a flowchart which shows the shape information calculation process of step S170 in the process shown in FIG. It is a flowchart which shows the inspection process of step S190 in the process shown in FIG. It is a functional block diagram which shows the structure of the processing part which concerns on 2nd Embodiment. It is a difference image obtained by mapping the difference value Δφ in two dimensions. It is a figure which shows the structure of the image inspection apparatus which concerns on 3rd Embodiment.

Preferred embodiments of the present invention will be described with reference to the accompanying drawings.

[First Embodiment]
FIG. 1 is a diagram showing a configuration of an image inspection device 1 according to the first embodiment. As shown in FIG. 1, the image inspection apparatus 1 according to the first embodiment mainly includes a projection unit 10, an imaging unit 20, and a processing unit 30.

The projection unit 10 is various known illuminations that irradiate light. In the first embodiment, the projection unit 10 is a flat plate-shaped illumination that irradiates diffused light. The projection unit 10 irradiates the object 40 with light having a striped pattern having a uniform pitch, and projects an image of the striped pattern onto the object 40. In the present embodiment, the projection unit 10 is described as irradiating light with a sine wave fringe pattern, but the present invention is not limited to this. Further, the pitch of the striped pattern can be appropriately changed according to the object 40 and the like.

The object 40 is an article to be inspected, and may be, for example, glass. Further, it is preferable that the surface of the object 40 does not include steep irregularities other than defects and has no pattern or the like.

The projection unit 10 irradiates light in a direction perpendicular to the surface of the flat plate (direction of arrow 14). At this time, when the direction parallel to the surface of the projection unit 10 (direction of arrow 12) and the direction in contact with the surface of the object 40 (direction of arrow 41) are parallel, a uniform pitch is applied to the surface of the object 40. The image of the striped pattern to have is projected. In the present embodiment, since the direction of the arrow 12 is inclined with respect to the direction of the arrow 41, the fringe pattern projected on the object 40 has a non-uniform pitch.

The photographing unit 20 is composed of various known cameras that capture images. The photographing unit 20 photographs an image of the striped pattern projected on the object 40. The photographing unit 20 transmits the captured image data (hereinafter, referred to as “image data”) to the processing unit 30.

FIG. 2 is a diagram showing an example of an image of a striped pattern photographed by the photographing unit 20. The striped pattern included in the image 500 shown in FIG. 2 changes in the direction indicated by the arrow 502. Moreover, the pitch of the striped pattern is non-uniform. Specifically, the pitch increases as it advances in the direction of arrow 502. For example, the pitch indicated by the arrow 508 on the right side is larger than the pitch indicated by the arrow 506 on the left side.

Further, on the surface of the object, there is a linear dent in the area 504 surrounded by the broken line. In the present embodiment, this dent will be described as a defect of the object.

The processing unit 30 calculates shape information using the image data acquired by the photographing unit 20, and inspects the object 40 based on the shape information. The function of the processing unit 30 will be described in detail with reference to FIG.

FIG. 3 is a functional block diagram showing an example of the configuration of the processing unit 30. As shown in FIG. 3, the processing unit 30 includes an image acquisition unit 310, a shape information calculation unit 320, an inspection unit 330, and a storage unit 340.

The image acquisition unit 310 acquires image data from the photographing unit 20. The image acquisition unit 310 transmits the acquired image data to the shape information calculation unit 320 and the storage unit 340.

The shape information calculation unit 320 calculates the shape information of the object 40 based on the image features extracted from the stripe pattern included in the image. The shape information is information regarding the shape of the surface of the object 40. Further, in the first embodiment, the image feature is the phase of the fringe pattern included in the image of the object. The function of the shape information calculation unit 320 will be described in more detail with reference to FIG.

FIG. 4 is a functional block diagram showing the configuration of the shape information calculation unit 320 according to the first embodiment. As shown in FIG. 4, the shape information calculation unit 320 includes a pitch estimation unit 322, a phase calculation unit 324, a phase connection unit 326, an angle calculation unit 328, and an image generation unit 329.

The pitch estimation unit 322 performs pitch estimation processing for estimating the pitch corresponding to each of the plurality of pixels included in the image. Specifically, the pitch estimation unit 322 first extracts one of the plurality of pixels. The pitch estimation unit 322 extracts a pixel group in which a plurality of pixels are arranged in a direction in which the stripe pattern changes (for example, in the left-right direction in the image 500 shown in FIG. 2) with the extracted pixels as the center. The pitch estimation unit 322 estimates the pitch of the stripes represented by the pixel values of this pixel group as the pitch corresponding to the extracted pixels. At this time, the pitch estimation unit 322 calculates the correlation coefficient between each of the plurality of sine waves having different pitches and the pixel group. The pitch estimation unit 322 estimates the pitch of the fringes represented by the pixel values of the pixel group as one of a plurality of sine waves according to the calculated correlation coefficient. The pitch estimation unit 322 generates a pitch distribution of an image by performing such a pitch estimation process for each of the plurality of pixels.

FIG. 5 is a diagram showing a pitch distribution image 510 generated by the pitch estimation unit 322 by performing a pitch estimation process on the image 500 shown in FIG. In the pitch distribution image 510, the larger the pitch, the whiter the pixels. Therefore, the pitch is larger toward the right side.

The phase calculation unit 324 performs a phase calculation process for calculating the phase corresponding to the pixels included in the image. Specifically, the phase calculation unit 324 extracts one of a plurality of pixels included in the pitch distribution image 510. The phase calculation unit 324 further extracts a pixel group in which a number of pixels corresponding to the estimated pitch corresponding to the extracted pixels are arranged in the direction in which the fringe pattern changes. The phase calculation unit 324 calculates the phase of a predetermined frequency component (for example, fundamental frequency) obtained by performing a discrete Fourier transform on the extracted pixel group. The phase calculation unit 324 generates a phase distribution by performing a phase calculation process for each of the plurality of pixels.

FIG. 6 is a diagram showing a first phase image 520 showing the phase distribution generated by the phase calculation unit 324 using the pitch distribution image 510 shown in FIG. The color density in the first phase image 520 indicates the magnitude of the phase (−π to + π). As shown in FIG. 6, when moving along the left-right direction (that is, the direction in which the fringe pattern changes), a boundary is created between the black part (phase: −π) and the white part (phase: + π). ing.

The phase connection unit 326 connects the phase boundaries in the first phase image 520. More specifically, the phase connection unit 326 connects the boundaries in which the phases are discontinuous by adding the phases by + 2π each time the phase changes from + π to −π. As a result, a second phase image 530 in which the phases are continuously mapped is generated.

FIG. 7 is a diagram showing a second phase image 530 generated by connecting the phases of the first phase image 520 shown in FIG. 6 by the phase connecting unit 326. The color intensity in the second phase image 530 indicates the magnitude of the phase. As shown in FIG. 7, in the second phase image 530, the phases are smoothly connected.

The angle calculation unit 328 converts the phase in the second phase image 530 into an angle described later. A method of calculating the angle by the angle calculation unit 328 will be described with reference to FIG.

Here, it is assumed that the projection unit 10 projects a sine wave fringe pattern having a pitch P onto the surface of the object. First, a case where an image of a striped pattern is projected on the surface 44 of a flat object having no unevenness will be described. At this time, the photographing unit 20 photographs an image of the striped pattern projected on the surface 44 of the object.

It is assumed that the optical axis 26 of the photographing unit 20 is specularly reflected by the surface 44 of the object, and the optical axis 18 reaches the point O of the projection unit 10. The distance from the lens 24 of the photographing unit 20 to the surface 44 of the object is W, and the distance from the surface 44 of the object to the point O of the projection unit 10 is V. The X-axis direction of the projection unit 10 is parallel to the plane of the projection unit 10 and is a direction in which the phase of the fringe pattern changes. The Z-axis direction of the projection unit 10 is a direction perpendicular to the plane of the projection unit 10. The optical axis 18 that is specularly reflected by the surface 44 of the object intersects the Z axis at an angle θ. Further, there is a u-axis on the element of the photographing unit 20, and the focal length of the photographing unit is f.

At this time, the relationship between the coordinates X of the projection unit 10 and the coordinates u of the photographing unit 20 is expressed by the following equation (1).

Here, assuming that the phase of the fringe pattern observed in the pixel of the coordinate u is φ, there is a relationship of X = φP / 2π between the coordinate X of the projection unit 10 and the phase φ. Substituting this relationship into equation (1) leads to the following equation (2).

As a method of obtaining the relationship between the coordinate u and the phase φ, there is a method of approximating by using the least squares method or the like from the combination of the plurality of observed coordinates u and the phase φ. When adopting the idea of this method, the phase φ is represented by the following equation (3) using the function of the fraction by the constants A, B and C.

Next, a case where a defect is generated on the surface of the object and the normal direction of the surface is tilted by an angle Ψ will be described. In this case, the direction of the optical axis 17 reflected by the surface 46 of the object is tilted by 2Ψ as compared with the case where the surface 44 is flat. As a result, the position of the fringe pattern observed at the coordinates u of the projection unit 10 is deviated by the distance dx as compared with the case where the surface 44 of the object is flat, and the observed phase φ'is the following equation (4). ).

Solving the above equation (4) for the angle Ψ leads to the following equation (5).

By observing the phase φ'of the fringe pattern at the coordinates u of the photographing unit 20 and using Eq. (5), the angle Ψ on the surface 46 of the defect of the object can be obtained.

In the present embodiment, the angle calculation unit 328 calculates the angle Ψ corresponding to each of the plurality of phases included in the second phase image 530 by the method described here. In this embodiment, it is assumed that the angle Ψ is the shape information.

The method by which the angle calculation unit 328 obtains the relationship between the coordinates u and the phase φ is not limited to the method described above, and may be a method obtained by inputting each parameter into the above equation (2).

The image generation unit 329 generates a shape image (first shape image) by mapping the calculated shape information in two dimensions. In the present embodiment, the image generation unit 329 generates a shape image by mapping the calculated angle Ψ in two dimensions.

FIG. 9 is a diagram showing a shape image 540 generated by two-dimensionally mapping the calculated angle Ψ by the image generation unit 329. In FIG. 9, the larger the angle Ψ, the whiter the pixels. In particular, the pixels are white in the area 542 surrounded by the broken line square. This region 542 corresponds to a dented portion included in the region 504 in the image 500 shown in FIG.

The inspection unit 330 inspects the object based on the shape information. In the present embodiment, first, the inspection unit 330 compares each of the angles Ψ included in the shape image with the threshold value T and binarizes them. Specifically, the inspection unit 330 generates a binary image by converting pixels having an angle Ψ of the threshold value T or more into white and pixels having an angle Ψ of less than the threshold value T of black.

FIG. 10 is a diagram showing a binary image 550 generated by the inspection unit 330 by binarizing the angle Ψ included in the shape image 540 shown in FIG. 9. As described above, the white region corresponds to the angle Ψ above the threshold T and the black region corresponds to the angle Ψ below the threshold T. The wider the white area, the more likely it is that the object has defects.

In this embodiment, the inspection unit 330 inspects the object based on the area of the white area. More specifically, the inspection unit 330 can compare the area of the white region with the threshold value and inspect the object based on the comparison result. For example, when the area of the white area is equal to or larger than the threshold value A, the inspection unit 330 determines that the object has a defect. On the other hand, when the area of the white region is less than the threshold value A, the inspection unit 330 can determine that there is no defect in the object.

The storage unit 340 stores various information related to processing by the processing unit 30. For example, the storage unit 340 stores various information such as image data acquired by the image acquisition unit 310, shape information calculated by the shape information calculation unit 320, and inspection results by the inspection unit 330. The information stored in the storage unit 340 is referred to by the shape information calculation unit 320 and the inspection unit 330 as necessary.

FIG. 11 is a diagram showing a physical configuration of the processing unit 30 according to the present embodiment. The processing unit 30 communicates with a CPU (Central Processing Unit) 30a corresponding to a calculation unit, a RAM (Random Access Memory) 30b corresponding to a storage unit 340, and a ROM (Read only Memory) 30c corresponding to a storage unit 340. It has a unit 30d, an input unit 30e, and a display unit 30f. Each of these configurations is connected to each other via a bus so that data can be transmitted and received. In this example, the case where the processing unit 30 is composed of one computer will be described, but the processing unit 30 may be realized by combining a plurality of computers. Further, the configuration shown in FIG. 11 is an example, and the processing unit 30 may have configurations other than these, or may not have a part of these configurations.

The CPU 30a is a control unit that controls execution of a program stored in the RAM 30b or ROM 30c, calculates data, and processes data. The CPU 30a is a calculation unit that executes a program (image inspection program) for inspecting the object based on the image of the object irradiated with the light of the striped pattern. The CPU 30a receives various data from the input unit 30e and the communication unit 30d, displays the calculation result of the data on the display unit 30f, and stores the data in the RAM 30b.

The RAM 30b is a storage unit 340 in which data can be rewritten, and may be composed of, for example, a semiconductor storage element. The RAM 30b may store data such as a program executed by the CPU 30a and an image of an object. It should be noted that these are examples, and data other than these may be stored in the RAM 30b, or a part of these may not be stored.

The ROM 30c can read data from the storage unit 340, and may be composed of, for example, a semiconductor storage element. The ROM 30c may store, for example, an image inspection program or data that is not rewritten.

The communication unit 30d is an interface for connecting the processing unit 30 to another device. The communication unit 30d may be connected to a communication network such as the Internet.

The input unit 30e receives data input from the user, and may include, for example, a keyboard and a touch panel.

The display unit 30f visually displays the calculation result by the CPU 30a, and may be configured by, for example, an LCD (Liquid Crystal Display). The display unit 30f may display an image of the captured object.

The image inspection program may be stored in a storage medium readable by a computer such as a RAM 30b or a ROM 30c and provided, or may be provided via a communication network connected by a communication unit 30d. In the processing unit 30, the CPU 30a executes the image inspection program to realize various operations described with reference to FIGS. 3 and 4. It should be noted that these physical configurations are examples and do not necessarily have to be independent configurations. For example, the processing unit 30 may include an LSI (Large-Scale Integration) in which the CPU 30a and the RAM 30b or ROM 30c are integrated.

The configuration and function of the image inspection device 1 according to the present embodiment have been described above. Next, the processing of the image inspection apparatus 1 according to the present embodiment will be described with reference to FIGS. 12 to 14. FIG. 12 is a flowchart showing the processing of the image inspection device 1 according to the present embodiment. Further, FIG. 13 is a flowchart showing the shape information calculation process of step S170 in the process shown in FIG. Further, FIG. 14 is a flowchart showing the inspection process of step S190 in the process shown in FIG. Hereinafter, the processing of the image inspection apparatus 1 will be described with reference to FIGS. 12 to 14.

First, the projection unit 10 irradiates the object 40 with light in a striped pattern (step S110). As a result, an image of the striped pattern is projected on the object. Next, the photographing unit 20 photographs an image of the striped pattern projected on the object 40 (step S130).

Next, the image acquisition unit 310 acquires image data from the photographing unit 20 (step S150). Next, the shape information calculation unit 320 performs a shape image generation process (step S170). Specifically, the processing unit 30 calculates shape information based on the image data and generates a shape image. The shape image generation process will be described in more detail with reference to FIG.

FIG. 12 is a flowchart showing the procedure of the shape information calculation process. First, the pitch estimation unit 322 performs pitch estimation processing on the image data to generate a pitch distribution image (step S171). Next, the phase calculation unit 324 generates a first phase image based on the pitch distribution image (step S173). Next, the phase connection unit 326 generates a second phase image by making a phase connection with respect to the first phase image (step S175). Next, the angle calculation unit 328 calculates the angle Ψ as shape information based on the second phase image (step S177). Next, the image generation unit 329 generates a shape image by mapping the calculated angle Ψ in two dimensions (step S179). When the shape image is generated, the shape information calculation process ends.

Returning to FIG. 12, the continuation of the processing of the image inspection apparatus 1 will be described. When the shape image generation process is performed in step S170, the inspection unit 330 performs the inspection process (step S190). The inspection process will be described in more detail with reference to FIG.

First, the inspection unit 330 generates a binary image formed by a white region and a black region, as described with reference to FIG. 10 (step S191). Next, the inspection unit 330 determines whether or not the area of the white region is equal to or larger than the threshold value (step S193). When it is determined that the area of the white region is equal to or larger than the threshold value (step S193: YES), the inspection unit 330 determines that the object is defective (step S195). On the other hand, when it is determined that the area of the white region is equal to or larger than the threshold value (step S193: NO), the inspection unit 330 determines that the object has no defect (step S197). When the inspection unit 330 determines the presence or absence of defects in the object, the process shown in FIG. 11 ends.

The first embodiment has been described above. According to the present embodiment, the projection unit 10 and the photographing unit 20 are arranged so that the fringe pattern included in the image projected on the object 40 has a non-uniform pitch. Therefore, the arrangement of the projection unit 10 and the photographing unit 20 becomes more free than in the case where the image of the striped pattern having a uniform pitch is projected on the object 40. As a result, it becomes convenient to arrange the projection unit 10 and the photographing unit 20 at desired positions according to the position where the object 40 is inspected. For example, the projection unit 10 can be arranged so that the direction parallel to the plane of the projection unit 10 (direction of arrow 12 in FIG. 1) and the direction parallel to the surface of the object 40 (direction of arrow 41 in FIG. 1) are different. Therefore, according to the image inspection apparatus 1 according to the present embodiment, the projection unit 10 can be arranged more freely as compared with the technique described in Patent Document 1.

Further, in the present embodiment, the object 40 can be inspected by using one image. Therefore, the object 40 can be inspected in a short time. Therefore, according to the image inspection device 1 according to the present embodiment, it is possible to inspect an object in a shorter time than when a plurality of images are taken.

Further, according to the image inspection apparatus 1 according to the present embodiment, it is possible to detect a gentle defect existing in the object 40.

In the present embodiment, the inspection unit 330 inspects the object by using the angle Ψ of the surface of the object as shape information. Not limited to this, the inspection unit 330 can inspect the object by using various shape information based on the image features extracted from the stripe pattern. Specifically, the inspection unit 330 can inspect the object by using various shape information based on the phase of the stripe pattern.

For example, the shape information calculation unit 320 differentiates the phase φ'(u) of the second phase distribution generated by the phase connection unit 326, and calculates the differential value dφ'(u) / du as the shape information. Can be done. For example, when the differential value dφ'(u) / du exceeding the threshold value is present, the inspection unit 330 can determine that the object is defective. Before differentiating the phase φ'(u), the shape information calculation unit 320 smoothes the second phase distribution with a spatial filter such as a Gaussian filter as preprocessing, and the phase in a region having a specific width. Differentiation of φ'(u) may be performed.

Further, as will be described later, the shape information calculation unit 320 can also calculate the difference value between the phase φ'(u) and the phase φ (u) of the fringe pattern projected on the flat surface as the shape information.

In the present embodiment, the image generation unit 329 has been described as generating a shape image by mapping the angle Ψ in two dimensions. Not limited to this, the image generation unit 329 may generate a shape image by applying a spatial filter to the angle Ψ. For example, the image generation unit 329 can selectively generate an image of a fringe pattern in a specific frequency band by using a spatial filter such as a Laplacian of Gaussian. When the shape information generated in this way exceeds the threshold value, the inspection unit 330 may determine that the object has a defect.

Further, the shape information calculation unit 320 utilizes the fact that the sign of the angle Ψ is opposite between the case where the surface of the object is convex and the case where the surface is concave, and determines the angle Ψ of either the convex surface or the concave surface. It may be acquired selectively.

Further, in the present embodiment, the inspection unit 330 has been described as determining the presence or absence of defects in the object. The inspection unit 330 may further detect the position of a defect (abnormality) existing in the object. For example, the inspection unit 330 may detect the position of the white region in the binary image 550 shown in FIG. 10 as the position of the defect.

[Second Embodiment]
In the second embodiment, the description of the matters common to the first embodiment will be omitted, and only the differences will be described. In the above embodiment, it has been described that the inspection unit 330 mainly inspects the object using the angle Ψ. In the second embodiment, the inspection unit 330 inspects the object by the shape information based on the phase generated by the phase connection unit 326 without using the angle Ψ.

FIG. 15 is a functional block diagram showing the configuration of the processing unit 31 according to the second embodiment. The processing unit 31 according to the second embodiment includes a correction unit 350 in addition to the image acquisition unit 310, the shape information calculation unit 320, the inspection unit 330, and the storage unit 340 included in the processing unit 30 according to the first embodiment.

The correction unit 350 corrects the misalignment of the image. More specifically, the correction unit 350 corrects the positional deviation of the first shape image (second phase image in this embodiment) generated by the shape information calculation unit 320 and the second shape image described later. As a result, even when the first shape image and the second shape image before correction are displaced, the inspection unit 330 can appropriately inspect the object.

In the second embodiment, a shape image (also referred to as a “second shape image”) that two-dimensionally represents the shape of an article that is known to be a non-defective product is prepared. The prepared second shape image is stored in the storage unit 340. In the present embodiment, the second shape image is an image generated by two-dimensionally mapping the phase φ of the fringe pattern when the fringe pattern is projected on a flat article without defects.

In the present embodiment, the inspection unit 330 uses the second phase image generated by the phase connection unit 326 as the first shape image. The inspection unit 330 inspects the object based on the second phase image and the second shape image. For example, the shape information calculation unit 320 calculates the difference value Δφ between the phase φ ′ of the second phase image and the phase φ of the second shape image. FIG. 16 is a difference image 560 obtained by mapping the difference value Δφ in two dimensions. The larger the difference value Δφ is, the whiter it is displayed. For example, when the difference image contains a difference value Δφ which is equal to or larger than the threshold value, the inspection unit 330 may determine that the object has no defect.

When the inspection unit 330 compares the first shape image with the second shape image, the correction unit 350 may correct the misalignment of the first shape image and the second shape image in advance. As a result, even when the first shape image and the second shape image before correction have a positional deviation, the inspection unit 330 can appropriately inspect the object.

The inspection unit 330 may project the first shape image as necessary. Assuming that the surface of the object is almost flat, it is also possible to estimate the projective transformation parameters of the fringe pattern. Examples of the method of estimating the projective transformation parameter include a method of superimposing a specific pattern on the projected fringe pattern and projecting the pattern. Further, the method of estimating the projective transformation parameter may be a method of using the shape information of the object (for example, the contour of the object).

The second embodiment has been described above.

(supplement)
In the present embodiment, the inspection unit 330 has a phase φ'of the second phase image generated by phase connection by the phase connection unit 326 and a phase φ when a striped pattern is projected on a flat article without defects. It was explained as the one to be inspected using the difference value with. Not limited to this, the inspection unit 330 uses the difference value between the first phase image before the phase connection by the phase connection unit 326 and the phase when the fringe pattern is projected on a flat article without defects. You may inspect.

In the present embodiment, the inspection unit 330 has been described as inspecting the object by using the difference value Δφ between the phase value φ'of the object and the phase value φ of the non-defective product. Not limited to this, the inspection unit 330 may inspect the object by using the difference between the angle Ψ of the surface of the object and the angle of the surface of the non-defective product.

Further, in the present embodiment, the inspection unit 330 has been described as inspecting the object by comparing the shape image of the article without defects with the shape image based on the image of the striped pattern projected on the object. Not limited to this, the inspection unit 330 may inspect the object by comparing the shape image of the defective article with the shape image based on the image of the striped pattern projected on the object.

[Third Embodiment]
FIG. 17 is a diagram showing a configuration of an image inspection device 2 according to a third embodiment. The image inspection device 2 shown in FIG. 17 is different from the image inspection device 1 shown in FIG. 1 and further includes a reflector 19. The projection unit 10 reflects the light of the striped pattern on the reflector 19 and projects the image of the striped pattern on the object 40. Also in this case, the projection unit 10 can project an image of a striped pattern having a non-uniform pitch on the surface of the object 40 via the reflector 19.

The embodiments described above are for facilitating the understanding of the present invention, and are not for limiting and interpreting the present invention. Each element included in the embodiment and its arrangement, material, condition, shape, size, etc. are not limited to those exemplified, and can be changed as appropriate. In addition, the configurations shown in different embodiments can be partially replaced or combined.

In the above embodiment, it has been described that the projection unit 10 mainly irradiates the object 40 with the light of one striped pattern. Not limited to this, the projection unit 10 may irradiate the object 40 with light having a plurality of stripe patterns in which the directions of the stripes are different from each other. For example, the projection unit 10 may sequentially irradiate the object 40 with light of two stripe patterns whose stripe directions differ from each other by 90 ° by, for example, a display capable of displaying an image. The shape information calculation unit 320 can calculate the shape information for each image of the plurality of stripe patterns and synthesize the calculated shape information of the plurality of stripe patterns. The inspection unit 330 can inspect the object 40 more accurately by inspecting the object 40 based on the synthesized shape information.

[Appendix]
A projection unit (10) that irradiates an object with light having a striped pattern having a uniform pitch and projects an image of the striped pattern onto the object.
An imaging unit (20) that captures an image of a striped pattern projected on the object, and
A shape information calculation unit (320) that calculates shape information of the object based on image features extracted from the stripe pattern included in the image, and
An inspection unit (330) that inspects the object based on the shape information is provided.
The striped pattern contained in the image has a non-uniform pitch.
Image inspection device (1, 2).

1,2 ... Image inspection device, 10 ... Projection unit, 20 ... Imaging unit, 30, 31 ... Processing unit, 40 ... Object, 310 ... Image acquisition unit, 320 ... Shape information calculation unit, 329 ... Image generation unit, 330 ... Inspection unit, 350 ... Correction unit, 500 ... Image, 520 ... First phase image, 530 ... Second phase image, 540 ... Shape image

Claims (9)

A projection unit that irradiates an object with light of a striped pattern having a uniform pitch and projects an image of the striped pattern onto the object.
An imaging unit that captures an image of the striped pattern projected on the object, and
A shape information calculation unit that calculates shape information of the object based on image features extracted from the fringe pattern included in the image, and a shape information calculation unit.
It is provided with an inspection unit that inspects the object based on the shape information.
The striped pattern contained in the image has a non-uniform pitch.
Image inspection equipment. The inspection unit detects the position of the abnormality in the object.
The image inspection apparatus according to claim 1. The shape information calculation unit calculates shape information corresponding to each of a plurality of positions in the striped pattern image, and generates a first shape image by mapping the calculated shape information in two dimensions.
The image inspection apparatus according to claim 1 or 2. The inspection unit detects an abnormality of the object based on the second shape image representing the shape of a specific article in two dimensions and the first shape image.
The image inspection apparatus according to claim 3. A correction unit for correcting the positional deviation of the first shape image and the second shape image is further provided.
The image inspection apparatus according to claim 4. The inspection unit binarizes the shape information included in the first shape image with a threshold value, and based on the area corresponding to any of the two values in the first shape image, the object is described. Inspect,
The image inspection apparatus according to claim 3. The projection unit is a flat plate-shaped illumination that irradiates diffused light, and irradiates the object with the light of the striped pattern.
The image inspection apparatus according to any one of claims 1 to 6. The projection unit irradiates the object with light having a plurality of stripe patterns having different stripe directions.
The photographing unit captures an image of each of the plurality of striped patterns.
The shape information calculation unit calculates shape information for each image of the plurality of stripe patterns, and synthesizes the calculated shape information of the plurality of stripe patterns.
The inspection unit inspects the object based on the synthesized shape information.
The image inspection apparatus according to any one of claims 1 to 7. By irradiating an object with light of a striped pattern having a uniform pitch and projecting an image of the striped pattern onto the object,
Taking an image of the striped pattern projected on the object and
To calculate the shape information of the object based on the image features extracted from the striped pattern in the image,
Inspecting the object based on the shape information, including
The striped pattern contained in the image has a non-uniform pitch.
Image inspection method.

Images