Inspection apparatus, inspection method, and program

Abstract

To inspect the appearance of a tire without irradiating the tire with band-like light or line-like light.SOLUTION: An inspection device includes shape recognition means for recognizing a shape of a tire from an image of the tire and detecting a defect candidate of the tire, tire information recognition means for extracting a character from the image of the tire and recognizing tire information including a model number of the tire, reference dimension setting means for acquiring information on a dimension in a specification of the tire based on the recognized model number of the tire, dimension calculation means for calculating a dimension of the defect candidate based on the dimension in the specification of the tire and the shape of the tire, and determination means for determining presence or absence of a defect of the tire based on the dimension of the defect candidate.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an inspection device, an inspection method, and a program. [Background technology]

[0002] One method for inspecting the appearance of tires is to use the light cutting method. For example, a surface shape measuring device described in Patent Document 1 irradiates a measurement surface of an object to be measured, such as a tire, with a band-like or line-like light, and sequentially captures light-section lines formed by the band-like or line-like light moving across the measurement surface as a rotary table rotates the object to be measured. This surface shape measuring device then generates an image showing the surface shape of the measurement surface by sequentially arranging multiple sets of light-section line image data according to the corresponding rotation angles. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-131124 Summary of the Invention [Problem to be solved by the invention]

[0004] When inspecting the appearance of a tire using the light-section method, it is necessary to irradiate each part of the tire with a strip-shaped or line-shaped light and capture images, which is thought to take a long time for the inspection. If it were possible to inspect the appearance of a tire without irradiating the tire with a strip-shaped or line-shaped light, it is expected that the time required for the inspection would be relatively short.

[0005] An example of an object of the present disclosure is to provide an inspection device, an inspection method, and a program that can solve the above-mentioned problems. [Means for solving the problem]

[0006] According to a first aspect of the present disclosure, an inspection device includes a shape recognition means that recognizes the shape of a tire from an image of the tire and detects candidate defects in the tire; a tire information recognition means that extracts characters from the image of the tire and recognizes tire information including the tire model number; a reference dimension setting means that acquires information on the tire's specification dimensions based on the recognized tire model number; a dimension calculation means that calculates the dimensions of the candidate defects based on the tire's specification dimensions and the tire shape; and a determination means that determines whether or not the tire has a defect based on the dimensions of the candidate defects.

[0007] According to a second aspect of the present disclosure, an inspection method includes a computer recognizing a shape of a tire from an image of the tire, detecting potential tire defects, extracting characters from the image of the tire to recognize tire information including the tire model number, obtaining information on tire specification dimensions based on the recognized tire model number, calculating dimensions of the potential defect based on the tire specification dimensions and the tire shape, and determining whether or not there is a defect in the tire based on the dimensions of the potential defect.

[0008] According to a third aspect of the present disclosure, a program causes a computer to perform the following operations: recognize the shape of a tire from an image of the tire and detect candidate defects in the tire; extract characters from the image of the tire to recognize tire information including the tire model number; obtain information on the tire's specification dimensions based on the recognized tire model number; calculate the dimensions of the candidate defects based on the tire's specification dimensions and the tire shape; and determine whether or not the tire has a defect based on the dimensions of the candidate defects. [Effects of the Invention]

[0009] According to one aspect of the present disclosure, the appearance of a tire can be inspected without the need to irradiate the tire with a strip or line of light. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an inspection apparatus according to at least one embodiment. [Figure 2] FIG. 1 illustrates an example of a camera arrangement according to at least one embodiment. [Figure 3] FIG. 1 illustrates an example of a camera arrangement according to at least one embodiment. [Figure 4] 1 is a diagram illustrating an example of a procedure for a process of tire inspection performed by an inspection device according to at least one embodiment. [Figure 5] FIG. 1 is a diagram illustrating an example of the configuration of an inspection apparatus according to at least one embodiment. [Figure 6] 1A to 1C are diagrams illustrating an example of a processing procedure in an inspection method according to at least one embodiment. [Figure 7] FIG. 1 illustrates an example configuration of a computer according to at least one embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] The following describes embodiments of the present invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0012] First Embodiment 1 is a diagram showing an example of the configuration of an inspection device according to at least one embodiment. In the configuration shown in FIG. 1, the inspection device 1 includes an imaging unit 100 and a processing unit 200. The photographing unit 100 includes a lighting device 110, a camera 120, a rangefinder 130, a photographing control unit 140, an operation mechanism 150, an operation control unit 160, and a synchronization unit 170. The operation mechanism 150 includes a roller 151 and a pusher 152.

[0013] The processing unit 200 includes a shape recognition unit 210 , a tire information recognition unit 220 , a reference dimension setting unit 230 , a tire information master unit 240 , a dimension calculation unit 250 , and a determination unit 260 . 1 also shows a tire 910 to be inspected. The tire 910 may be configured external to the inspection device 1.

[0014] The lighting device 110 irradiates the tire with light (illumination light) for photographing the tire. The lighting device 110 is an example of a lighting means. The camera 120 photographs the tire. The distance meter 130 measures the distance between the distance meter 130 itself and the tire 910. The distance measured by the distance meter 130 is used to calculate the outer diameter of the tire 910. The outer diameter of the tire 910 is used to calculate the amount of rotation of the roller 151 that rotates the tire 910.

[0015] The photographing control unit 140 controls the illumination of the lighting device 110 and the photographing of the tire 910 by the camera 120 . For example, the photography control unit 140 may switch the lighting pattern of the lighting device 110 for each part of the tire 910, causing the lighting device 110 to irradiate the part with light from various directions. Then, the photography control unit 140 may cause the camera 120 to take an image for each lighting pattern.

[0016] Here, it is assumed that the direction of light irradiation onto the tire 910 differs for each illumination pattern of the lighting device 110. For example, the lighting device 110 may include multiple light sources, and the light source that is turned on may be changed for each illumination pattern. The lighting device 110 irradiates light onto the tire 910 from light sources at different positions for each illumination pattern, so that the direction of light irradiation differs for each illumination pattern.

[0017] It is expected that the processing unit 200 will be able to recognize the shape of the tire 910 in more detail if the camera 120 captures an image of the tire 910 in each of a plurality of light irradiation directions. For example, if the camera 120 captures an image of a portion of the tire 910 in each of three or more light irradiation directions, the processing unit 200 will be able to recognize the shape of the tire 910 using a photometric stereo method.

[0018] The operation mechanism 150 performs an operation on the tire 910 . The rollers 151 rotate the tire 910 so that the camera 120 can photograph various parts of the tire 910 . The roller 151 is an example of a rotating means. The pusher 152 pushes open the tire 910 from the inside, which is expected to make it easier to detect cracks in the tire, if any.

[0019] The operation control unit 160 controls the operation mechanism 150 to perform an operation on the tire 910. In particular, the operation control unit 160 calculates the outer diameter of the tire 910 based on the distance measured by the rangefinder 130, and calculates the amount of rotation of the roller 151 required to rotate the tire 910 once (one revolution) based on the calculated outer diameter. Then, the operation control unit 160 rotates the roller 151 by a predetermined amount each time the camera 120 captures an image of the tire 910. The operation control unit 160 repeats the rotation of the tire 910 by the rotation of the roller 151 until the tire 910 completes one revolution. This allows the camera 120 to capture images of the entire circumference of the tire 910 without missing any. Furthermore, the operation control unit 160 controls the pusher 152 so that the pusher 152 presses the tire 910 against the roller and spreads the tire 910 from the inside.

[0020] The synchronization unit 170 synchronizes the photographing of the tire 910 by the camera 120 with the operation of the operation mechanism 150 on the tire 910 . Specifically, the synchronization unit 170 instructs the imaging control unit 140 to have the camera 120 capture an image of the tire 910. When the camera 120 has completed capturing an image of a portion of the tire 910, the synchronization unit 170 instructs the operation control unit 160 to rotate the tire 910. The synchronization unit 170 repeats the instruction to capture an image of the tire 910 and the instruction to rotate the tire 910 until the entire circumference of the tire 910 has been captured.

[0021] The shape recognition unit 210 recognizes the shape of the tire 910 using the image captured by the camera 120 . The shape recognition unit 210 is an example of a shape recognition means. In the following, an example will be described in which the shape recognition unit 210 uses a photometric stereo method to recognize the shape of the tire 910. However, the method by which the shape recognition unit 210 recognizes the shape of the tire 910 is not limited to a specific method.

[0022] The tire information recognition unit 220 reads information about the tire 910 from the image captured by the camera 120. Specifically, the tire information recognition unit 220 extracts text information from the image of the side of the tire 910 and reads tire information including the model number of the tire 910. The tire information here refers to information related to the tire 910. The tire information recognition unit 220 corresponds to an example of tire information recognition means. The method by which the tire information recognition unit 220 reads characters from an image is not limited to a specific method, and the tire information recognition unit 220 may read characters from an image using a known character recognition algorithm.

[0023] The tire information master unit 240 stores tire information in advance (before the inspection device 1 starts inspection). In particular, the tire information master unit 240 stores tire information that links the tire model number with the tire's standard dimensions. The standard dimensions here are dimensions that are determined as specifications.

[0024] The reference dimension setting unit 230 acquires information on the reference dimensions of the tire 910 based on the model number of the tire 910 recognized by the tire information recognition unit 220. Specifically, the reference dimension setting unit 230 acquires information on the reference dimensions linked to the model number of the tire 910 from the tire information master unit 240. The reference dimension setting unit 230 corresponds to an example of a reference dimension setting means.

[0025] The dimension calculation unit 250 calculates the absolute dimensions of the tire 910 based on the shape of the tire 910 recognized by the shape recognition unit 210 and the reference dimensions of the tire 910 acquired by the reference dimension setting unit 230. In particular, the dimension calculation unit 250 calculates the dimensions of the defect candidates of the tire 910 (absolute dimensions of the defect candidates of the tire 910). The absolute dimensions here refer to the actual dimensions (real dimensions) of the tire 910. The dimension calculation unit 250 corresponds to an example of a dimension calculation means.

[0026] For example, the shape recognition unit 210 may use a photometric stereo method to generate an image in which the contrast of grooves (grooves on the tread surface) is emphasized. Then, the dimension calculation unit 250 may calculate how many pixels each groove corresponds to, and calculate how many millimeters one pixel corresponds to based on the groove dimension information indicated in the reference dimensions. The dimension calculation section 250 can convert the number of pixels in the image of the tire 910 into the absolute dimension of the tire 910 based on the calculated length per pixel. However, the method by which the dimension calculation section 250 calculates the absolute dimension of the tire 910 is not limited to a specific method.

[0027] The determination unit 260 determines whether or not there is a defect in the tire 910 based on the tire information of the tire 910 recognized by the tire information recognition unit 220, the shape of the tire 910 recognized by the shape recognition unit 210, and the dimensions of the tire 910 calculated by the dimension calculation unit 250. The determination unit 260 is an example of a determination means.

[0028] For example, if the shape recognition unit 210 detects a defect candidate such as a crack in the recognized shape of the tire 910, the dimension calculation unit 250 calculates the absolute dimension of the defect candidate. Then, the determination unit 260 compares the absolute dimension of the defect candidate with a threshold value set for each type of defect. If the absolute dimension of the defect candidate is greater than or equal to the threshold value, the determination unit 260 determines that the defect candidate corresponds to a defect. Furthermore, if the magnitude of the error between the absolute dimension of the tire 910 calculated by the dimension calculation unit 250 and the reference dimension of the tire 910 acquired by the reference dimension setting unit 230 is greater than or equal to a predetermined threshold value, the determination unit 260 determines that the tire 910 has a defect.

[0029] The types of defects detected by the determination unit 260 are not limited to a specific type. For example, the determination unit 260 may detect tread wear (insufficient groove depth due to wear of the tread portion), cuts, cracks, penetrations (nail holes, etc.), ripples, curling, exposed internal cords (metal cords or nylon cords inside the tire), and puncture repair marks, or some of these, but are not limited to these.

[0030] The shape recognition unit 210 may also use information on the reference dimensions of the tire 910 when detecting defect candidates. For example, the information on the reference dimensions of the tire 910 may include information on the grooves (tread pattern) of the tire 910 in the specifications. The shape recognition unit 210 may then refer to the information on the grooves of the tire 910 to distinguish between cuts, scratches, cracks, and grooves. This is expected to enable the shape recognition unit 210 to detect defect candidates with higher accuracy.

[0031] Fig. 2 is a diagram showing an example of the arrangement of the camera 120. Fig. 2 shows an example of the arrangement of the camera 120 when the tire 910 is viewed from the tread surface side. 2, the inspection device 1 is equipped with three cameras 120. When distinguishing between the three cameras 120, they are also referred to as cameras 120-1, 120-2, and 120-3. The camera 120-1 captures an image of the surface of the tire 910 from outside the tire 910. For example, the camera 120-1 may capture an image of the tread portion and shoulder portion of the tire 910. The camera 120-1 is an example of a tread surface camera that captures an image of the tread surface side of a tire.

[0032] The camera 120-2 captures an image of the side surface of the tire 910 from outside the tire 910. For example, the camera 120-2 may capture an image of the bead portion, the sidewall portion, and the shoulder portion of the tire 910. The camera 120-2 is an example of a side camera that captures images of the side of a tire.

[0033] The camera 120-3 captures an image of the inner surface of the tire 910. For example, the camera 120-3 may capture an image of the belt, carcass, and beads inside the tire 910. The camera 120-3 is an example of an internal camera that captures the inside of a tire.

[0034] Fig. 3 is a diagram showing an example of the arrangement of the camera 120. Fig. 3 shows an example of the arrangement of the camera 120 in the example of Fig. 2 when the tire 910 is viewed from the side. Fig. 3 also shows an example of the arrangement of the rangefinder 130, roller 151, and pusher 152. Two rollers 151 are shown in Fig. 3. When distinguishing between the two rollers 151, they are also referred to as rollers 151a and 151b.

[0035] 2, the camera 120-1 captures an image of the surface of the tire 910 from outside the tire 910. The camera 120-2 captures an image of the side surface of the tire 910 from outside the tire 910. The camera 120-3 captures an image of the inside surface of the tire 910.

[0036] Further, the distance meter 130 is disposed so as to face in a direction perpendicular to the tread surface of the tire 910, and measures the distance to the tread surface of the tire 910. The outer diameter of the tire 910 can be calculated based on the distance measured by the distance meter 130. The amount of rotation of the roller 151 that rotates the tire 910 can be calculated based on the outer diameter of the tire. However, the rangefinder 130 is not essential for the inspection device 1. For example, the operation control unit 160 may refer to an image of the side surface of the tire 910 captured by the camera 120-2 and detect that the roller 151 has made one rotation of the tire 910. 3, two rollers 151 support the upright tire 910 from below. However, the tire 910 may also be placed in a horizontal position.

[0037] The cameras 120-1 and 120-3 may each capture an image of half of the tire 910 in the width direction. In this case, when the roller 151 rotates the tire 910 once, the cameras 120-1, 120-2, and 120-3 capture an image of one circumference of half the tire 910 (an image of the entire circumference). When the tire 910 is placed upside down and the roller 151 rotates the tire 910 once more, the cameras 120-1, 120-2, and 120-3 can capture an image of the entire tire 910.

[0038] In this way, by turning the tire 910 over, it is possible to capture an image of the entire tire 910 using a relatively small number of cameras 120. The tire 910 may be turned over manually or by the operation mechanism 150.

[0039] Alternatively, the imaging unit 100 may be provided with six cameras, so that the camera 120 can simultaneously capture an image of the half of the tire 910 opposite to the half captured by the camera 120 in the examples of Figures 2 and 3. In this case, if the roller 151 rotates the tire 910 once, an image of the entire tire 910 can be captured without having to turn the tire 910 over.

[0040] Alternatively, a person may visually inspect a portion of the tire 910. For example, in the examples of Figures 2 and 3, the camera 120-3 may not be provided, and instead an inspection worker may visually check the inside of the tire 910.

[0041] Furthermore, the wavelength of light captured by the camera 120 is not limited to a specific wavelength, and may be various wavelengths depending on the wavelength of light emitted by the lighting device 110. For example, the lighting device 110 may emit near infrared (NIR) light, and the camera 120 may be a near infrared camera. When the photographing section 100 includes a plurality of cameras 120, the wavelength (frequency range) of light photographed by the cameras 120 may be the same for all the cameras 120, or may be different for each camera 120.

[0042] Alternatively, one camera 120 may move to capture images of multiple locations on the tire 910. For example, one camera 120 may capture images from the position and direction of camera 120-1 in the example of FIGS. 2 and 3, and roller 151 may rotate the tire 910 one revolution. Next, camera 120 may capture images from the position and direction of camera 120-2 in the example of FIGS. 2 and 3, and roller 151 may rotate the tire 910 one revolution. Then, camera 120 may capture images from the position and direction of camera 120-3 in the example of FIGS. 2 and 3, and roller 151 may rotate the tire 910 one revolution. Furthermore, the tire 910 may be positioned by determining whether it is on the front or back, and camera 120 and roller 151 may similarly capture images of one revolution of the tire 910 from each of the positions of camera 120-1, camera 120-2, and camera 120-3.

[0043] In this way, when camera 120 moves to capture images, the mechanism for moving camera 120 is not limited to a specific mechanism. For example, camera 120 may be attached to a robot arm (articulated arm), and imaging control unit 140 may move camera 120 by operating the robot arm, but is not limited to this.

[0044] FIG. 4 is a diagram showing an example of a procedure for the inspection device 1 to inspect a tire. 4, the camera 120 captures an image of the tire 910 (step S101). When the shape recognition unit 210 uses the photometric stereo method, the lighting device 110 switches between three or more lighting patterns under the control of the photography control unit 140. Then, the camera 120 captures an image of the same location on the tire 910 for each lighting pattern under the control of the photography control unit 140.

[0045] Next, the roller 151 rotates the tire 910 by a predetermined amount (step S102). After step S102, the process returns to step S101. The inspection device 1 repeatedly performs the process of step S101 and the process of step S102 to photograph each part of the tire 910.

[0046] 2, the camera 120 may capture images of half of the tire 910 over the entire circumference while the roller 151 rotates the tire 910 once. The tire 910 may then be placed upside down, and the camera 120 may capture images of the other half of the tire 910 over the entire circumference while the roller 151 rotates the tire 910 once.

[0047] When the entire tire 910 has been photographed, the shape recognition unit 210 uses the image data obtained by the photography to generate information indicating the appearance of the tire 910 (step S111). Here, the information indicating the appearance of the tire 910 is information indicating at least one of the shape of the tire 910, the pattern of the tire 910, and the characters printed on the tire 910.

[0048] For example, the shape recognition unit 210 may use a photometric stereo method to generate normal information and contrast-enhanced images of the outer surface, side surface, and inner surface of the tire 910. The normal information here is information that indicates the shape of the tire 910 using normal vectors relative to the surface of the tire 910. The normal information is an example of three-dimensional shape data. The contrast-enhanced image here is an image in which the contrast is enhanced from a captured image of the tire 910. The contrast-enhanced image is an example of two-dimensional image data.

[0049] Next, the shape recognition unit 210 uses the normal information and / or the contrast-enhanced image to detect defect candidates in the tire 910 and calculates position information of the detected defect candidates (step S112). Information indicating the defect candidates and their positions is also referred to as defect candidate information. The shape recognition unit 210 may accumulate the defect candidate information in a database of defect candidate information.

[0050] Furthermore, the tire information recognition unit 220 determines whether or not there is text on the side of the tire 910 based on the captured image of the side of the tire 910 or the contrast-enhanced image of the tire 910 generated by the shape recognition unit 210 in step S111, and reads the text if it is determined that there is text (step S121). The captured image or contrast-enhanced image of the tire 910 is an example of an image of the tire 910.

[0051] There is no particular limitation on the method by which the tire information recognition unit 220 reads characters from the photographed image or contrast-enhanced image of the side of the tire 910. For example, the tire information recognition unit 220 may read characters using a known character recognition method, but is not limited to this.

[0052] If the characters are read in step S121, the tire information recognition unit 220 estimates the model number of the tire 910 from the character reading result (step S122). For example, the tire information master unit 240 may use the tire name and size information shown on the side of the tire as the tire model number, and store information linking the tire model number with the standard dimensions. Then, the tire information recognition unit 220 may extract the tire name and size designation from the character string read in step S121.

[0053] The method by which the tire information recognition unit 220 extracts the tire name and size designation from the character string is not limited to a specific method. For example, the tire information recognition unit 220 may use a known character string search method to extract, from the character string of the read result, a character string that matches one of the multiple tire names stored in the tire information master unit 240. Similarly, the tire information recognition unit 220 may extract, from the character string of the read result, a character string that matches one of the multiple model number designations stored in the tire information master 204.

[0054] Next, the reference dimension setting unit 230 acquires the information on the reference dimensions of the tire stored in the tire information master unit 240, which is linked to the tire name and size notation extracted by the tire information recognition unit 220 (step S123).

[0055] Next, the dimension calculation unit 250 calculates the absolute dimensions of the tire 910 based on the reference dimensions set by the reference dimension setting unit 230 and the information indicating the appearance of the tire 910 generated by the shape recognition unit 210 (step S131). In particular, the dimension calculation unit 250 calculates the absolute dimensions of the defect candidates detected by the shape recognition unit 210. If the shape recognition unit 210 detects multiple defect candidates in step S112, the dimension calculation unit 250 calculates the dimensions of each defect candidate.

[0056] Next, the determination unit 260 determines whether or not there is a defect in the tire 910 based on the dimensions of the defect candidate calculated by the dimension calculation unit 250 (step S132). For example, the determination unit 260 may compare the dimensions of the defect candidate calculated by the dimension calculation unit 250 with a threshold value that is predetermined for each type of defect. Then, if the dimensions of the defect candidate are greater than or equal to the threshold value, the determination unit 260 may determine that the defect candidate corresponds to a defect. After step S132, the inspection device 1 ends the process of FIG.

[0057] If the tire information recognition unit 220 is unable to estimate the model number of the tire 910 in the processing of steps S121 and S122, the dimension calculation unit 250 may calculate the absolute dimensions of the tire 910 based on information indicating the appearance, without using the reference dimensions, in step S131. Alternatively, the tire information recognition unit 220 may notify the user that the model number of the tire 910 could not be estimated, and may accept input of model number information by the user.

[0058] As described above, the shape recognition unit 210 recognizes the shape of the tire 910 from the image of the tire 910 and detects possible defects in the tire 910. The tire information recognition unit 220 extracts characters from the image of the tire 910 and recognizes tire information including the tire model number. The reference dimension setting unit 230 acquires information on the tire's specification dimensions based on the recognized tire model number. The dimension calculation unit 250 calculates the dimensions of the defect candidate based on the specified dimensions of the tire 910 and the shape of the tire 910. The determining unit 260 determines whether or not the tire 910 has a defect based on the dimensions of the defect candidate.

[0059] The inspection device 1 makes it possible to inspect the appearance of a tire without the need to irradiate the tire with a strip-shaped or line-shaped light. Because black objects like tires do not reflect light well, inspection using images captured by a camera is generally difficult. Therefore, inspections are typically performed by irradiating the tire with a strip or line of light. However, when irradiating a tire with a strip or line of light, the area of ​​the tire that can be illuminated with light per irradiation is relatively small, and it may take a long time to inspect the entire circumference of the tire. Furthermore, when irradiating a tire with a strip or line of light, it is necessary to strictly synchronize the moving speed of the tire surface with the scan rate of the imaging unit in order to accurately measure dimensions, especially in the feed direction. This requires the time and effort of resetting the settings to match the diameter and width of each tire size.

[0060] In contrast, the inspection device 1 illuminates a single inspection point from multiple directions and captures an image, making it possible to extract with high precision changes in reflected light due to the minute surface shapes, even for objects that do not easily reflect light, such as tires. Also, capturing an image of a surface rather than a band (line) enables accurate inspection even without strict synchronization with the moving speed of the tire surface.

[0061] Furthermore, because there are many different types of tires with different diameters and widths, it is considered difficult to determine accurate dimensions from a photographed image of the tire alone. Even if tire inspection could be automated, manually setting inspection parameters for each tire type and diameter would be a burden on the inspector.

[0062] In contrast, in the inspection device 1, the tire information recognition unit 220 reads the model number from the image of the tire 910, and the reference dimension setting unit 230 acquires information on the reference dimensions linked to the model number, thereby automatically acquiring parameters for calculating absolute dimensions from the captured image based on the reference dimensions. The inspection device 1 eliminates the need to manually set parameters for each tire type and diameter, thereby reducing the burden on the operator performing the inspection.

[0063] Furthermore, the camera 120-1 (tread surface camera), the camera 120-2 (side camera), and the camera 120-3 (internal camera) capture images of the tread surface side, the side side, and the interior of the tire 910, respectively. The determining unit 260 determines whether or not there are defects on the tread surface side, the side surfaces, and the interior of the tire 910. The inspection device 1 can inspect the tread surface side, side surface side, and interior of the tire 910. Furthermore, an image captured by the camera 120-2 or an image obtained by enhancing the contrast of that image can be used as an image for estimating the tire model number.

[0064] The roller 151 also rotates the tire 910 around the rotation axis of the tire 910 . Each of the camera 120-1 (tread surface camera), camera 120-2 (side camera), and camera 120-3 (internal camera) takes an image every time the tire 910 rotates, and this process is repeated until the entire circumference of the tire 910 has been photographed. The inspection device 1 can inspect the tread surface side, side surfaces, and interior of the tire 910 over the entire circumference of the tire 910.

[0065] In addition, each of camera 120-1 (tread surface camera), camera 120-2 (side camera), and camera 120-3 (internal camera) is positioned and oriented so as to capture one of the two areas that divide tire 910 at its center in the width direction. The rollers 151 rotate the tire 910 once each on the front and back sides of the tire 910 that are placed upside down. According to the inspection device 1, the entire tire 910 can be inspected.

[0066] Furthermore, the camera 120 is provided so that its position and orientation can be changed. The roller 151 rotates the tire 910 once for each position and orientation of the camera 120 . According to the inspection device 1, a relatively wide area of ​​the tire 910 can be inspected with one camera 120. For example, according to the inspection device 1, the entire tire 910 can be inspected with one camera 120.

[0067] The distance meter 130 is provided at a position and in a direction that allows it to measure the distance from the tread surface of the tire in a direction perpendicular to the tread surface. The roller 151 rotates the tire 910 by an amount calculated based on the distance measured by the distance meter 130 as the amount of rotation of the tire 910 for one rotation.

[0068] According to the inspection device 1, it is possible to prevent or reduce excessive or insufficient rotation of the tire 910 when inspecting the entire circumference of the tire 910. Furthermore, since the inspection device 1 can obtain data for one circumference of the tire 910, it is relatively easy to associate the obtained data with the position of the tire 910.

[0069] Furthermore, the lighting device 110 can emit light in each of a plurality of irradiation directions. The shape recognition unit 210 applies the photometric stereo method to images obtained by the camera 120 capturing images of the same location on the tire 910 from each of a plurality of irradiation directions, and generates information indicating the shape of the tire 910. According to the inspection device 1, the shape of the tire 910 can be detected with a relatively high degree of accuracy using a photometric stereo method.

[0070] Furthermore, the shape recognition unit 210 generates three-dimensional shape data and two-dimensional image data using the photometric stereo method. According to the inspection device 1, the dimension calculation unit 250 can calculate the absolute dimensions of the tire 910 with a relatively high degree of accuracy using three-dimensional shape data obtained by photometric stereo and two-dimensional image data. In particular, according to the inspection device 1, it is possible to calculate the dimensions of defect candidates with a relatively high degree of accuracy, and it is expected that the determination unit 260 can determine the presence or absence of a defect with a relatively high degree of accuracy.

[0071] Second Embodiment 5 is a diagram illustrating an example of the configuration of an inspection device according to at least one embodiment. In the configuration illustrated in FIG. 5, an inspection device 610 includes a shape recognition unit 611, a tire information recognition unit 612, a reference dimension setting unit 613, a dimension calculation unit 614, and a determination unit 615.

[0072] With this configuration, the shape recognition unit 611 recognizes the shape of the tire from the tire image and detects possible tire defects. The tire information recognition unit 612 extracts characters from the image of the tire and recognizes tire information including the tire model number. The reference dimension setting unit 613 acquires information on the tire's specified dimensions based on the recognized tire model number. The dimension calculation unit 614 calculates the dimensions of the defect candidate based on the tire's specified dimensions and the tire's shape. The determining unit 615 determines whether or not there is a defect in the tire based on the dimensions of the defect candidate.

[0073] The shape recognition unit 611 corresponds to an example of a shape recognition means. The tire information recognition unit 612 corresponds to an example of a tire information recognition means. The reference dimension setting unit 613 corresponds to an example of a reference dimension setting means. The dimension calculation unit 614 corresponds to an example of a dimension calculation means. The determination unit 615 corresponds to an example of a determination means.

[0074] The inspection device 610 allows the appearance of a tire to be inspected without the need to irradiate the tire with a strip or line of light. However, because black objects such as tires do not reflect light well, it is generally difficult to inspect them using images taken with a camera, and inspections are usually performed by shining a strip or line of light onto the tire. However, when inspecting tires by shining a strip or line of light onto them, the area of ​​the tire that can be illuminated with light in each irradiation is relatively small, and it can take a long time to inspect the entire circumference of the tire.

[0075] In contrast to this, the inspection device 610 can use information on the reference dimensions (information on the dimensions in the specifications) to inspect the tire with a relatively high degree of accuracy using images taken by a camera. For example, the dimension calculation unit 614 is expected to be able to calculate the dimensions of the defect candidate with a relatively high degree of accuracy because it calculates the dimensions of the defect candidate based on the reference dimensions. The dimension calculation unit 614's ability to calculate the dimensions of the defect candidate with a relatively high degree of accuracy allows the determination unit 615 to determine the presence or absence of a defect with a relatively high degree of accuracy.

[0076] Furthermore, because there are many different types of tires and different diameters, it is considered difficult to determine accurate dimensions from a photographed image of the tire alone. Even if tire inspection could be automated, manually setting inspection parameters for each tire type and diameter would be a burden on the inspector.

[0077] In contrast, in the inspection device 610, the tire information recognition unit 612 reads the model number from the tire image, and the reference dimension setting unit 613 acquires the information on the reference dimensions linked to the model number, thereby enabling the dimensions of the reference information to be acquired automatically. The inspection device 1 eliminates the need to manually set each tire type and diameter, thereby reducing the burden on the operator performing the inspection.

[0078] <Third embodiment> FIG. 6 is a diagram showing an example of a processing procedure in an inspection method according to at least one embodiment. The inspection method shown in FIG. 6 includes recognizing the shape (step S611), recognizing tire information (step S612), and setting reference dimensions (step S613). This includes calculating the dimensions (step S614) and making a determination (step S615).

[0079] In recognizing the shape (step S611), a computer recognizes the shape of the tire from an image of the tire and detects possible defects in the tire. In recognizing tire information (step S612), the computer extracts characters from the tire image and recognizes tire information including the tire model number. In setting the reference dimensions (step S613), the computer obtains information on the tire's specification dimensions based on the recognized tire model number. In calculating dimensions (step S614), a computer calculates dimensions of the defect candidate based on the tire's specification dimensions and the tire's shape. In making a determination (step S615), the computer determines whether or not the tire has a defect based on the dimensions of the defect candidate.

[0080] According to the inspection method shown in FIG. 6, the appearance of a tire can be inspected without the need to irradiate the tire with a strip-shaped or line-shaped light. However, because black objects such as tires do not reflect light well, it is generally difficult to inspect them using images taken with a camera, and inspections are usually performed by shining a strip or line of light onto the tire. However, when inspecting tires by shining a strip or line of light onto them, the area of ​​the tire that can be illuminated with light in each irradiation is relatively small, and it can take a long time to inspect the entire circumference of the tire.

[0081] In contrast to this, the inspection method shown in FIG. 6 uses information on standard dimensions (information on dimensions according to specifications) and can therefore perform tire inspection with a relatively high degree of accuracy using images captured by a camera. For example, in calculating the dimensions (step S614), the computer calculates the dimensions of the defect candidate based on the reference dimensions, and it is expected that the dimensions of the defect candidate can be calculated with a relatively high degree of accuracy. Since the dimensions of the defect candidate can be calculated with a relatively high degree of accuracy in calculating the dimensions (step S614), the computer can determine with a relatively high degree of accuracy whether a defect exists or not in making the determination (step S615).

[0082] Furthermore, because there are many different types of tires and different diameters, it is considered difficult to determine accurate dimensions from a photographed image of the tire alone. Even if tire inspection could be automated, manually setting inspection parameters for each tire type and diameter would be a burden on the inspector.

[0083] In contrast, in the inspection method shown in Fig. 6, tire information is recognized (step S612), and the computer reads the model number from the tire image, and in setting the reference dimensions (step S613), the computer obtains the information on the reference dimensions linked to the model number, thereby automatically obtaining the dimensions of the reference information. According to the inspection method shown in Fig. 6, there is no need to manually set each tire type and diameter, and the burden on the worker performing the inspection can be relatively reduced.

[0084] FIG. 7 illustrates an example configuration of a computer according to at least one embodiment. In the configuration shown in FIG. 7, a computer 700 includes a CPU 710, a main memory device 720, an auxiliary memory device 730, an interface 740, and a non-volatile recording medium 750.

[0085] One or more of the above-described inspection apparatus 1 and inspection apparatus 610, or a part thereof, may be implemented in a computer 700. In this case, the operation of each of the above-described processing units is stored in the auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from the auxiliary storage device 730, loads it into the main storage device 720, and executes the above-described processing in accordance with the program. The CPU 710 also allocates storage areas in the main storage device 720 corresponding to each of the above-described storage units in accordance with the program. Communication between each device and other devices is performed by an interface 740 having a communication function and performing communication under the control of the CPU 710. The interface 740 also has a port for a nonvolatile recording medium 750, and reads information from the nonvolatile recording medium 750 and writes information to the nonvolatile recording medium 750.

[0086] When the inspection device 1 is implemented in the computer 700, the operations of the photography control unit 140, operation control unit 160, synchronization unit 170, shape recognition unit 210, tire information recognition unit 220, reference dimension setting unit 230, dimension calculation unit 250, and determination unit 260 are stored in the form of a program in the auxiliary storage device 730. The CPU 710 reads the program from the auxiliary storage device 730, loads it into the main storage device 720, and executes the above-mentioned processing in accordance with the program.

[0087] Furthermore, the CPU 710, in accordance with the program, allocates storage areas in the main storage device 720 for the inspection device 1 to perform processing, such as a storage area for the tire information master unit 240. Communication between the inspection device 1 and other devices is performed by the interface 740, which has a communication function and operates under the control of the CPU 710. Interaction between the inspection device 1 and a user is performed by the interface 740, which has an input device and an output device, presenting information to the user via the output device under the control of the CPU 710 and accepting user operations via the input device.

[0088] When the inspection device 610 is implemented in the computer 700, the operations of the shape recognition unit 611, the tire information recognition unit 612, the reference dimension setting unit 613, the dimension calculation unit 614, and the determination unit 615 are stored in the form of a program in the auxiliary storage device 730. The CPU 710 reads the program from the auxiliary storage device 730, loads it into the main storage device 720, and executes the above-mentioned processing in accordance with the program.

[0089] Furthermore, the CPU 710 allocates a storage area in the main storage device 720 for the inspection device 610 to perform processing in accordance with the program. Communication between the inspection device 610 and other devices is performed by the interface 740, which has a communication function and operates under the control of the CPU 710. Interaction between the inspection device 610 and a user is performed by the interface 740, which has an input device and an output device, presenting information to the user via the output device under the control of the CPU 710 and accepting user operations via the input device.

[0090] One or more of the above-described programs may be recorded on nonvolatile recording medium 750. In this case, interface 740 may read the programs from nonvolatile recording medium 750. CPU 710 may then directly execute the programs read by interface 740, or may temporarily store the programs in main storage device 720 or auxiliary storage device 730 and then execute them.

[0091] Note that the processing of each part may be performed by recording a program for executing all or part of the processing performed by the inspection device 1 and the inspection device 610 on a computer-readable recording medium, and having a computer system read and execute the program recorded on the recording medium. Note that the "computer system" here includes the OS (Operating System) and hardware such as peripheral devices. Furthermore, "computer-readable recording media" refers to portable media such as flexible disks, optical magnetic disks, ROMs (Read Only Memory), and CD-ROMs (Compact Disc Read Only Memory), as well as storage devices such as hard disks built into computer systems. The program may be one that realizes part of the aforementioned functions, or may be one that can realize the aforementioned functions in combination with a program already stored in the computer system.

[0092] Although the embodiments have been described above, the specific configuration is not limited to these embodiments, and the present invention also includes designs that do not deviate from the gist of the present invention. Furthermore, the above-described embodiments can be combined with other embodiments as appropriate.

[0093] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0094] (Appendix 1) a shape recognition means for recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; a tire information recognition means for extracting characters from the tire image and recognizing tire information including the tire model number; a reference dimension setting means for acquiring information on the tire's specification dimensions based on the recognized tire model number; a size calculation means for calculating the size of the defect candidate based on the tire's specified size and the tire's shape; a determination means for determining whether or not there is a defect in the tire based on the dimensions of the defect candidate; An inspection device comprising:

[0095] (Appendix 2) A tread surface camera, a side camera, and an internal camera for photographing the tread surface side, side side, and internal side of the tire, respectively. Equipped with The determination means determines the presence or absence of defects on the tread surface side, the side surface side, and the interior of the tire. 10. The inspection device described in Appendix 1.

[0096] (Appendix 3) a rotation means for rotating the tire around a rotation axis, The tread surface camera, the side camera, and the internal camera each take an image every time the tire rotates, and this process is repeated until the entire circumference of the tire is photographed. 1. The inspection device described in Appendix 2.

[0097] (Appendix 4) The tread surface camera, the side camera, and the internal camera are each disposed at a position and orientation capable of photographing one of two regions obtained by dividing the tire at the center in the width direction thereof, The rotating means rotates the tire once before and after the tire is placed upside down. 1. The inspection device described in Appendix 3.

[0098] (Appendix 5) a camera whose position and orientation are variable; a rotation means for rotating the tire once for each position and orientation of the camera; 2. The inspection device according to claim 1, comprising:

[0099] (Appendix 6) a distance meter provided at a position and in a direction to measure a distance from the tread surface of the tire in a direction perpendicular to the tread surface, the rotation means rotates the tire by an amount of rotation calculated based on the distance measured by the distance meter as the amount of rotation for one rotation of the tire; 6. The inspection device according to any one of appendices 3 to 5.

[0100] (Appendix 7) an illumination means capable of irradiating light in each of a plurality of irradiation directions; Equipped with the shape recognition means applies a photometric stereo method to images obtained by taking images of the same location on the tire with a camera in each of the plurality of irradiation directions, thereby generating information indicating the shape of the tire. 7. The inspection device of any one of appendices 2 to 6.

[0101] (Appendix 8) the shape recognition means generates three-dimensional shape data and two-dimensional image data using the photometric stereo method; 8. The inspection device described in Appendix 7.

[0102] (Appendix 9) The computer Recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; Extracting characters from the image of the tire and recognizing tire information including the tire model number; Based on the recognized tire model number, obtain the tire specification dimension information; Calculating the size of the defect candidate based on the tire's specification size and the tire's shape; determining whether or not there is a defect in the tire based on the dimensions of the defect candidate; An inspection method including:

[0103] (Appendix 10) In the determination, the computer determines the presence or absence of defects on the tread surface side, side surface side, and interior of the tire based on images taken by a tread surface camera, a side camera, and an interior camera, which take images of the tread surface side, side surface side, and interior of the tire, respectively. Testing method described in Appendix 9.

[0104] (Appendix 11) The computer causing each of the tread surface camera, the side camera, and the internal camera to take an image each time the tire is rotated by a rotation means that rotates the tire around a rotation axis, until the entire circumference of the tire is photographed; 11. The testing method of claim 10, comprising:

[0105] (Appendix 12) The tread surface camera, the side camera, and the internal camera are each disposed at a position and orientation capable of photographing one of two regions obtained by dividing the tire at the center in the width direction thereof, The computer causes the rotation means to rotate the tire once before and after the tire is placed upside down. The test method described in Appendix 11.

[0106] (Appendix 13) the computer causes the rotation means to rotate the tire once for each position and orientation of a camera provided so as to be variable in position and orientation; Testing method described in Appendix 9.

[0107] (Appendix 14) The computer The rotation means is caused to rotate the tire by an amount calculated based on a distance measured by a rangefinder provided at a position and in a direction that measures the distance from the direction perpendicular to the tread surface of the tire, as the amount of rotation for one rotation of the tire. 14. The testing method of any one of appendices 11 to 13.

[0108] (Appendix 15) In recognizing the shape of the tire from the image of the tire, the computer applies a photometric stereo method to images obtained by a camera photographing the same location on the tire in each of a plurality of irradiation directions using lighting means capable of irradiating light in each of the plurality of irradiation directions, thereby generating information indicating the shape of the tire. 15. The testing method of any one of appendices 10 to 14.

[0109] (Appendix 16) In recognizing the shape of the tire from the image of the tire, the computer generates three-dimensional shape data and two-dimensional image data using the photometric stereo method. The test method described in Appendix 15.

[0110] (Appendix 17) On the computer, Recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; extracting characters from the image of the tire to recognize tire information including the tire model number; Obtaining tire specification dimension information based on the recognized tire model number; calculating a size of the defect candidate based on the tire's specified size and the tire's shape; determining whether or not the tire has a defect based on the dimensions of the defect candidate; A program that executes the following.

[0111] (Appendix 18) In the determination, the computer is caused to determine the presence or absence of defects on the tread surface side, the side surface side, and the interior of the tire based on images taken by a tread surface camera, a side camera, and an interior camera, which take images of the tread surface side, the side surface side, and the interior of the tire, respectively. 17. The program described in Appendix 17.

[0112] (Appendix 19) The computer, causing the tread surface camera, the side camera, and the internal camera to take images each time the tire is rotated by a rotation means that rotates the tire around a rotation axis, and repeating this process until images are taken around the entire circumference of the tire; 19. The program according to claim 18, which causes the program to execute the above.

[0113] (Appendix 20) The tread surface camera, the side camera, and the internal camera are each disposed at a position and orientation capable of photographing one of two regions obtained by dividing the tire at the center in the width direction thereof, The computer, causing the rotating means to rotate the tire once before and after the tire is placed upside down; 19. The program of claim 19,

[0114] (Appendix 21) The computer, causing a rotation means to rotate the tire once for each position and orientation of a camera provided so as to be variable in position and orientation; 21. The program of claim 20,

[0115] (Appendix 22) The computer, causing the rotation means to rotate the tire by an amount of rotation calculated based on a distance measured by a rangefinder provided at a position and in a direction that measures a distance from a direction perpendicular to the tread surface of the tire, as the amount of rotation for one rotation of the tire; 22. The program according to any one of appendices 19 to 21,

[0116] (Appendix 23) In recognizing the shape of the tire from the image of the tire, the computer is caused to apply a photometric stereo method to images obtained by a camera photographing the same location of the tire in each of a plurality of irradiation directions using lighting means capable of irradiating light in each of the plurality of irradiation directions, thereby generating information indicating the shape of the tire. 23. The program of any one of appendices 18 to 22.

[0117] (Appendix 24) In the step of recognizing the shape of the tire from the image of the tire, the computer is caused to generate three-dimensional shape data and two-dimensional image data by the photometric stereo method. 23. The program described in Appendix 23. [Explanation of symbols]

[0118] 1,610 Inspection equipment 100 Filming Department 110 Lighting equipment 120 Camera 130 Rangefinder 140 Imaging control unit 150 Operating mechanism 151 Laura 152 Pusher 160 Operation control section 170 Synchronization Unit 200 Processing section 210, 611 Shape recognition section 220, 612 Tire information recognition unit 230, 613 Reference dimension setting section 240 Tire Information Master Section 250, 614 Dimension calculation section 260, 615 Judgment section

Claims

1. a shape recognition means for recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; a tire information recognition means for extracting characters from the tire image and recognizing tire information including the tire model number; a reference dimension setting means for acquiring information on the tire's specification dimensions based on the recognized tire model number; a size calculation means for calculating the size of the defect candidate based on the tire's specified size and the tire's shape; a determination means for determining whether or not there is a defect in the tire based on the dimensions of the defect candidate; An inspection device comprising:

2. A tread surface camera, a side camera, and an internal camera for photographing the tread surface side, side side, and internal side of the tire, respectively. Equipped with The determination means determines the presence or absence of defects on the tread surface side, the side surface side, and the interior of the tire. The inspection device according to claim 1 .

3. a rotation means for rotating the tire around a rotation axis, The tread surface camera, the side camera, and the internal camera each take an image every time the tire rotates, and this process is repeated until the entire circumference of the tire is photographed. The inspection device according to claim 2 .

4. The tread surface camera, the side camera, and the internal camera are each disposed at a position and orientation capable of photographing one of two regions obtained by dividing the tire at the center in the width direction thereof, The rotating means rotates the tire once before and after the tire is placed upside down. The inspection device according to claim 3 .

5. a camera whose position and orientation are variable; a rotation means for rotating the tire once for each position and orientation of the camera; The inspection device according to claim 1 , comprising:

6. a distance meter provided at a position and in a direction to measure a distance from the tread surface of the tire in a direction perpendicular to the tread surface, the rotation means rotates the tire by an amount of rotation calculated based on the distance measured by the distance meter as the amount of rotation for one rotation of the tire; The inspection device according to claim 3 .

7. an illumination means capable of irradiating light in each of a plurality of irradiation directions; Equipped with the shape recognition means applies a photometric stereo method to images obtained by taking images of the same location on the tire with a camera in each of the plurality of irradiation directions, thereby generating information indicating the shape of the tire. The inspection device according to claim 2 .

8. the shape recognition means generates three-dimensional shape data and two-dimensional image data by the photometric stereo method; The inspection device according to claim 7.

9. The computer Recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; Extracting characters from the image of the tire and recognizing tire information including the tire model number; Based on the recognized tire model number, obtain the tire specification dimension information; Calculating the size of the defect candidate based on the tire's specification size and the tire's shape; determining whether or not there is a defect in the tire based on the dimensions of the defect candidate; An inspection method including:

10. On the computer, Recognizing the shape of the tire from an image of the tire and detecting possible defects in the tire; extracting characters from the image of the tire to recognize tire information including the tire model number; Obtaining tire specification dimension information based on the recognized tire model number; calculating a size of the defect candidate based on the tire's specified size and the tire's shape; determining whether or not the tire has a defect based on the dimensions of the defect candidate; A program that executes the following.

Images