Tire inspection support device, method, and program

Abstract

A tire inspection support device (1) comprises: a first detecting unit (11) for detecting a first region including a candidate of a cracked part of a tire, by means of first image processing with respect to a captured image of the tire; a generating unit (12) for generating a processing target image obtained by extracting a partial image corresponding to the first region from the captured image; a second detecting unit (13) for detecting a second region representing a cracked part, within the first region, by means of second image processing with respect to the processing target image; and a calculating unit (14) for calculating an index value relating to cracking of the tire, on the basis of the second region.

Description

[Technical Field]

[0001] The present invention relates to a tire inspection assistance device, method, and program, and more particularly to a tire inspection assistance device, method, and program for assisting tire inspection. [Background technology]

[0002] The criteria for determining whether tire maintenance is necessary vary depending on the driver and mechanic, and an objective indicator is needed. Patent Document 1 discloses a technology that extracts the groove area of ​​a tire from a distance image of the tire, detects cracks by binarizing the captured image of that area, and evaluates tire deterioration from the crack area. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-161575 Summary of the Invention [Problem to be solved by the invention]

[0004] Here, the degree of cracking must be taken into consideration when determining tire deterioration. Furthermore, there is room for improvement in the accuracy of technology for detecting cracks from captured images of tires. Furthermore, the technology disclosed in Patent Document 1 does not detect cracks other than those in grooves.

[0005] In view of the above-mentioned problems, an object of the present disclosure is to provide a tire inspection support device, method, and program for improving the accuracy of detecting cracks in captured images of a tire. [Means for solving the problem]

[0006] A tire inspection support device according to a first aspect of the present disclosure includes: a first detection means for detecting a first region including a candidate crack portion of the tire by performing a first image processing on a captured image of the tire; a generating means for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection means for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation means for calculating an index value related to cracks in the tire based on the second region; Equipped with.

[0007] A tire inspection support method according to a second aspect of the present disclosure includes: The computer Detecting a first region including a candidate crack portion of the tire by performing a first image processing on the captured image of the tire; generating a processing target image by extracting a partial image corresponding to the first region from the captured image; detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; An index value relating to cracks in the tire is calculated based on the second region.

[0008] A non-transitory computer-readable medium storing a tire inspection assistance program according to a third aspect of the present disclosure includes: a first detection process for detecting a first region including a candidate crack portion of the tire by performing a first image process on a captured image of the tire; a generation process for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection process for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation process for calculating an index value related to cracks in the tire based on the second region; to be executed by the computer. [Effects of the Invention]

[0009] The present disclosure makes it possible to provide a tire inspection support device, method, and program for improving the accuracy of detecting cracks from captured images of a tire. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing the configuration of a tire inspection support device according to a first embodiment. [Figure 2] 3 is a flowchart showing the flow of a tire inspection support method according to the first embodiment. [Figure 3] FIG. 10 is a block diagram showing the configuration of a tire inspection support system according to a second embodiment. [Figure 4] FIG. 10 is a block diagram showing the configuration of a tire inspection support device according to a second embodiment. [Figure 5] 10 is a flowchart showing the flow of a tire inspection support process according to the second embodiment. [Figure 6] 10 is a flowchart showing the flow of a tire inspection support process according to the second embodiment. [Figure 7] FIG. 10 is a diagram showing an example of a captured image (divided captured image) of a tire according to the second embodiment. [Figure 8] FIG. 10 is a diagram showing an example of a first binary image that is a result of detecting candidates for crack portions using a first image processing model according to the second embodiment. [Figure 9] FIG. 10 is a diagram showing an example of a processing target image according to the second embodiment. [Figure 10] FIG. 10 is a diagram showing an example in which a processing target image according to the second embodiment is divided into a plurality of first unit regions. [Figure 11] FIG. 10 is a diagram showing an example of a second binary image according to the second embodiment. [Figure 12] FIG. 10 is a diagram showing an example of a sliding window of a second binary image according to the second embodiment. [Figure 13] FIG. 10 is a diagram showing an example of a captured image (display image) to which the crack rate, the deterioration degree, and display information according to the second embodiment have been added. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are designated by the same reference numerals, and for clarity of explanation, duplicate explanations will be omitted as necessary.

[0012] <Embodiment 1> 1 is a block diagram showing the configuration of a tire inspection support device 1 according to the present embodiment 1. The tire inspection support device 1 is an information processing device for detecting cracked portions from captured images of a tire and supporting tire inspection. The tire inspection support device 1 includes a first detection unit 11, a generation unit 12, a second detection unit 13, and a calculation unit 14.

[0013] The first detection unit 11 performs first image processing on a captured image of the tire to detect a first region including candidates for cracks in the tire. The generation unit 12 generates a processing target image by extracting a partial image corresponding to the first region from the captured image. The second detection unit 13 performs second image processing on the processing target image to detect a second region indicating the crack from within the first region. The calculation unit 14 calculates an index value for tire cracks based on the second region.

[0014] 2 is a flowchart showing the flow of the tire inspection support method according to the first embodiment. First, the first detection unit 11 performs a first image processing on a captured image of the tire to detect a first region including candidates for cracked portions of the tire (S11). Next, the generation unit 12 generates a processing target image by extracting a partial image corresponding to the first region from the captured image (S12). Then, the second detection unit 13 performs a second image processing on the processing target image to detect a second region indicating a cracked portion from the first region (S13). Thereafter, the calculation unit 14 calculates an index value for cracks in the tire based on the second region (S14).

[0015] In this manner, the tire inspection support device 1 according to this embodiment uses a first image processing to detect a first region containing a candidate crack portion with a certain degree of accuracy (first accuracy) from a captured image of a tire. Then, the tire inspection support device 1 performs a second image processing on a partial image of the captured image corresponding to the first region. In other words, the second image processing narrows down the image data for crack detection. Therefore, the tire inspection support device 1 can more accurately detect a crack portion (second region) with a second accuracy higher than the first accuracy through the second image processing. Then, the tire inspection support device 1 calculates an index value from the second region indicating a more accurate crack portion according to a predetermined standard, thereby providing a user such as a mechanic with an objective index for determining whether tire maintenance is required. This improves the accuracy of detecting crack portions from captured images of a tire, thereby assisting tire inspections.

[0016] The tire inspection support device 1 includes a processor, a memory, and a storage device (not shown). The storage device stores a computer program that implements the processing of the tire inspection support method according to this embodiment. The processor then loads the computer program from the storage device into the memory and executes the computer program. This allows the processor to implement the functions of a first detection unit 11, a generation unit 12, a second detection unit 13, and a calculation unit 14.

[0017] Alternatively, each component of the tire inspection support device 1 may be realized by dedicated hardware. Furthermore, some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and programs. Furthermore, a CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field-Programmable Gate Array), quantum processor (quantum computer control chip), etc., may be used as the processor.

[0018] <Embodiment 2> The second embodiment is a specific example of the first embodiment described above. The problem to be solved by this embodiment will now be described in detail. First, tire wear can be classified into wear due to abrasion and wear due to cracking. The degree of wear due to abrasion is measured using a specific index value, namely, tire tread depth. Since a tire will not pass inspection unless its tread depth meets a threshold, mechanics can determine whether tire maintenance is required based on the objective index of tread depth. On the other hand, wear due to cracking occurs due to the influence of ultraviolet rays and the like, regardless of mileage, making it difficult for average drivers to predict. Furthermore, there are no clear maintenance standards for the degree of wear due to cracking. Therefore, the criteria for determining whether tire maintenance is required due to wear due to cracking vary among mechanics.

[0019] One way to address this issue is to use an AI (Artificial Intelligence) model that uses photographed images of tires as input to determine the degree of cracking in the tires. In this case, the AI ​​model is trained by deep learning or other methods using photographed images of the tires and the results of wear assessments by mechanics, etc. as learning data, and the trained model can then extract candidate images of cracked areas from the input data of photographed images of the tires.

[0020] However, in a trained model, the determination area of ​​a cracked portion may be determined (detected) as being larger than the actual area. As a result, there is a problem that the trained model alone cannot extract (detect) the detailed area and shape of the cracked portion. Therefore, in the following embodiments, a technique for improving the accuracy of detecting cracked portions from captured images of a tire will be described.

[0021] FIG. 3 is a block diagram showing the configuration of a tire inspection support system 1000 according to the second embodiment. The tire inspection support system 1000 is an information system for supporting the inspection of a tire 100 performed by a mechanic U. The tire inspection support system 1000 includes a camera 200, a tire inspection support device 300, and a display device 400. The camera 200 and the tire inspection support device 300 are connected via a communication line. Furthermore, the tire inspection support device 300 and the display device 400 are connected via a communication line. Here, the communication line may be a wired or wireless communication line or communication network. Examples of the communication line include a LAN (Local Area Network), the Internet, a wireless communication network, a mobile phone network, etc. Furthermore, the communication network may use any type of communication protocol.

[0022] The camera 200 is an imaging device used by the mechanic U to capture an image of the tire 100 to be inspected. The camera 200 transmits the captured image of the tire 100 to the tire inspection support device 300 by operation of the mechanic U or the like.

[0023] The tire inspection support device 300 is an example of the above-mentioned tire inspection support device 1. The tire inspection support device 300 may be configured redundantly with multiple servers, and each functional block may be realized by multiple computers.

[0024] The tire inspection support device 300 detects cracks in two stages from the captured image of the tire 100 captured by the camera 200 and calculates an index value for the cracked portion. The tire inspection support device 300 may also determine the degree of tire deterioration from the index value. The tire inspection support device 300 may also generate a display image by adding display information to the captured image for identifiably displaying positions corresponding to cracked portions in the captured image. Furthermore, the tire inspection support device 300 may output at least one of the index value, the deterioration degree determination result, and the display image to the display device 400.

[0025] The display device 400 displays on its screen at least one of the information received from the tire inspection support device 300, such as the index value, the result of the assessment of the degree of deterioration, and the display image. This makes it easier for the mechanic U to determine whether maintenance of the tire 100 is required via the screen of the display device 400.

[0026] FIG. 4 is a block diagram showing the configuration of a tire inspection support device 300 according to the second embodiment. The tire inspection support device 300 includes a storage unit 310, a memory 320, an IF (Interface) unit 330, and a control unit 340. The storage unit 310 is an example of a storage device such as a hard disk or a flash memory. The storage unit 310 stores a tire inspection support program 311 and a first image processing model 312. The tire inspection support program 311 is a computer program that implements the tire inspection support processing according to the second embodiment. The first image processing model 312 is a module for realizing the first image processing, and corresponds to the trained model described above.

[0027] The memory 320 is a volatile storage device such as a RAM (Random Access Memory), and is a storage area for temporarily storing information during operation of the control unit 340. The IF unit 330 is a communication interface between the inside of the tire inspection support device 300 and the camera 200 and display device 400.

[0028] The control unit 340 is a processor, i.e., a control device, that controls each component of the tire inspection support device 300. The control unit 340 loads the tire inspection support program 311 and the first image processing model 312 from the storage unit 310 into the memory 320, and executes the tire inspection support program 311 and the first image processing model 312. In this way, the control unit 340 realizes the functions of an acquisition unit 341, a preprocessing unit 342, a first detection unit 343, a generation unit 344, a second detection unit 345, a calculation unit 346, an identification unit 347, a determination unit 348, and an output unit 349.

[0029] The acquisition unit 341 acquires a captured image of the tire 100 from the camera 200. That is, the acquisition unit 341 accepts the captured image as an input to the tire inspection assistance program 311.

[0030] The pre-processing unit 342 converts the size (number of pixels, etc.) of the captured image into a first size, and divides the converted image into a plurality of divided captured images of each second size.

[0031] The first detection unit 343 is an example of the above-mentioned first detection unit 11. The first detection unit 343 performs first image processing on the captured image of the tire. Specifically, the first detection unit 343 performs first image processing on the captured image (segmented captured image) input to the tire inspection assistance program 311 using the first image processing model 312, and acquires a detection result. The detection result is information indicating a first region including a candidate crack portion of the tire. The detection result is, for example, rendering data of the region determined to be a candidate crack portion, or a first binary image in which the candidate crack portion is colored black and the other portions are colored white.

[0032] The generating unit 344 is an example of the generating unit 12 described above. The generating unit 344 generates a processing target image by extracting a partial image corresponding to the first region from the captured image. Specifically, the generating unit 344 extracts the partial image by masking regions other than the first region from the captured image, thereby generating the processing target image. The generating unit 344 may also generate a display image by adding display information of a position identified by the identifying unit 347 (described later) to the captured image. Here, the display information of the identified position is a frame surrounding the relevant region in the captured image, or information that highlights the relevant region so that it can be identified, or the like.

[0033] The second detection unit 345 is an example of the second detection unit 13 described above. The second detection unit 345 performs second image processing on the processing target image. Specifically, the second detection unit 345 performs binarization processing on the processing target image to generate a second binary image. Note that the second image processing is not limited to binarization processing. For example, the second image processing may be a process of extracting cracked portions using an AI model different from the first image processing, particularly a trained model. Furthermore, the second detection unit 345 may divide the processing target image into multiple first unit areas and perform second image processing for each first unit area to detect a second area for each first unit area.

[0034] The calculation unit 346 is an example of the calculation unit 14 described above. The calculation unit 346 calculates an index value related to cracks in the tire 100 based on the second region. For example, the calculation unit 346 may set a part of the processing target image as a second unit region, set multiple second unit regions within the processing target image by sliding the second unit region at a size narrower than the size of the second unit region, and calculate an index value for each second unit region. The size of the second unit region may be, for example, the vertical length or horizontal length of the second unit region. A size narrower than the size of the second unit region may be referred to as the sliding width. In other words, the calculation unit 346 sets multiple second unit regions by sliding the second unit region from the initial setting position within the processing target image by the sliding width so as to cover all regions within the processing target image. Therefore, the multiple second unit regions may overlap. The calculation unit 346 may then calculate, as the index value, a crack rate based on the ratio between the second unit region and the second region detected in the second unit region.

[0035] The identification unit 347 is an example of a first identification means. The identification unit 347 identifies a second unit area corresponding to one of the index values ​​calculated for each second unit area that satisfies a predetermined condition. The predetermined condition may be, for example, that the index value is equal to or greater than a first predetermined value, that the index value is the maximum value among the index values ​​calculated for each second unit area, or that the index value is in the upper rank or within a predetermined number of the index values. The identification unit 347 also identifies a position in the captured image that corresponds to the identified second unit area.

[0036] The determination unit 348 determines the degree of deterioration of the tire 100 based on the index value. For example, the determination unit 348 determines the degree of deterioration based on the crack rate. The degree of deterioration may be, for example, a multi-level value such as large, medium, or small, a numerical value, a percentage, or the like, but is not limited to these. Note that the determination unit 348 may determine the degree of deterioration for each second unit area.

[0037] The output unit 349 outputs the degree of deterioration determined by the determination unit 348 to the display device 400. The output unit 349 may also output the degree of deterioration for each second unit area. The output unit 349 may also output an index value. Specifically, the output unit 349 outputs the crack rate calculated by the calculation unit 346. The output unit 349 may also output the crack rate for each second unit area. The output unit 349 may add display information for identifiably displaying the position identified by the identification unit 347 to the captured image and output the image. Alternatively, the output unit 349 may output the display image generated by the generation unit 344. Alternatively, the output unit 349 may output the degree of deterioration and the index value in association with each other. Alternatively, the output unit 349 may output the degree of deterioration and the display information in association with each other. Alternatively, the output unit 349 may output the index value and the display information in association with each other. Alternatively, the output unit 349 may output the degree of deterioration, the index value, and the display information in association with each other.

[0038] 5 and 6 are flowcharts showing the flow of the tire inspection support process according to the present embodiment 2. First, a mechanic U photographs the tire 100 to be inspected using the camera 200. The camera 200 transmits the photographed image of the tire 100 to the tire inspection support device 300. In response, the tire inspection support device 300 acquires the photographed image of the tire 100 (S101).

[0039] Next, the tire inspection support device 300 converts the captured image to a first size and divides the converted image into a plurality of divided captured images of a second size (S102). For example, the pre-processing unit 342 converts the size of the captured image to 3000 pixels. Then, the pre-processing unit 342 divides the converted image into images of 500 x 500 pixels. However, conversion to the first size is not necessarily required. FIG. 7 is a diagram showing an example of a captured image of a tire (divided captured image 51) according to the second embodiment.

[0040] Next, the tire inspection support device 300 inputs each divided captured image into the first image processing model 312, and obtains the detection result of the candidate crack portion (first binary image) (S103).

[0041] FIG. 8 is a diagram showing an example of a first binary image that is a detection result 52 of candidate crack areas by the first image processing model 312 according to the second embodiment. The first image processing model 312 outputs the presence or absence of a possibility of a crack area for each pixel in the input dividedly captured image 51 as the detection result 52. The detection result 52 is an example of a binary image in which pixels determined to have a certain probability or higher of being a crack area are colored black, and pixels determined to have a probability lower than the certain probability are colored white. The set of black pixels in the detection result 52 is the first region described above. Note that the detection result 52 covers the actual crack area, but some areas other than cracks are also determined to have the possibility. In other words, the detection result 52 includes more candidate crack areas than actually exists.

[0042] Then, the tire inspection support device 300 generates a processing target image 53 from the divided captured image 51 and the detection result 52 (S104). For example, the generation unit 344 overlays the divided captured image 51 and the detection result 52, and converts pixels in the divided captured image 51 that correspond to white pixels in the detection result 52 to white. In other words, the pixel values ​​of pixels in the divided captured image 51 that correspond to black pixels in the detection result 52 are left as they are, and other pixels in the divided captured image 51 are masked to white. Note that in the above process, converting pixels in the divided captured image 51 to white is just one example, and any other conversion may be performed so that areas in the processing target image 53 that are not subject to the second image processing can be identified. For example, the generation unit 344 may perform conversion so that the opacity of areas in the divided captured image 51 that are not subject to the second image processing is 0, that is, transparent.

[0043] 9 is a diagram showing an example of a processing target image 53 according to the second embodiment. The processing target image 53 can be said to be obtained by replacing black pixels in the detection result 52 with corresponding pixels in the divided captured image 51, i.e., with partial images. The areas of the processing target image 53 that are partial images have different shading and colors. On the other hand, the areas of the processing target image 53 other than the partial images are white.

[0044] Then, the tire inspection support device 300 divides the processing target image 53 into a plurality of first unit areas (S105). For example, the second detection unit 345 divides the processing target image 53 into a 500 x 500 pixel grid. In other words, the second detection unit 345 divides the processing target image 53 into a plurality of first unit areas.

[0045] 10 is a diagram showing an example in which a processing target image 53 according to the second embodiment is divided into a plurality of first unit areas. For example, the processing target image 53 is shown divided into first unit areas 531, 532, 533, .... In this example, each first unit area has the same size, but the second detection unit 345 may divide the image into areas (grids) of different sizes. Furthermore, when dividing the image, the second detection unit 345 may add margins so that no portion is less than the first unit area.

[0046] Then, the tire inspection support device 300 performs binarization processing for each first unit area to detect cracks (S106). That is, the tire inspection support device 300 acquires a second binary image after the binarization processing for the processing target image 53. Specifically, the second detection unit 345 performs binarization processing for each first unit area using a predetermined binarization processing. For example, the second detection unit 345 converts each pixel in the first unit area 531 into black if the pixel value is equal to or greater than a threshold, and into white if the pixel value is less than the threshold. Note that the predetermined binarization processing may change the threshold for each first unit area.

[0047] FIG. 11 is a diagram showing an example of a second binary image 54 according to the second embodiment. In the second binary image 54, a portion of the partial image (first region) of the above-described processing target image 53 is converted to black (second region) and the remainder is converted to white. The group of black pixels in the second binary image 54 can be said to be the detection result of the second region. Furthermore, the second binary image 54 is white because the portion of the processing target image 53 other than the partial image (other than the first region) is originally white. Therefore, the second binary image 54 has fewer black regions (second regions) compared to the first region of the detection result 52 (first binary image).

[0048] Thereafter, the tire inspection support device 300 calculates the crack rate for each second unit area (S107). Note that the size of the second unit area in step S107 may be different from the size of the first unit area in step S105 described above. Furthermore, the tire inspection support device 300 may calculate the crack rate for each second unit area while shifting a second unit area of ​​a predetermined size by a predetermined number of pixels on the second binary image 54. The second unit area set in step S107 while shifting by a predetermined number of pixels in this manner is called a sliding window, or simply a window.

[0049] 12 is a diagram showing an example of a sliding window of the second binary image 54 according to the second embodiment. Here, the size of the sliding windows 541 to 543, etc., is assumed to be 600 pixels high and 600 pixels wide. The sliding width is assumed to be 200 pixels. In other words, the sliding window is shifted by 200 pixels at a time.

[0050] For example, the calculation unit 346 sets a sliding window 541 with the upper left corner of the second binary image 54 as the origin. Then, the calculation unit 346 calculates the crack rate of the sliding window 541. Specifically, the calculation unit 346 calculates the crack rate by dividing the number of black pixels in the sliding window 541 by the size of the sliding window. Next, the calculation unit 346 slides the left edge of the sliding window 541 200 pixels to the right to set the sliding window 542. As a result, a region 400 pixels wide from the right edge of the sliding window 541 overlaps with a region 400 pixels wide from the left edge of the sliding window 542. Then, the calculation unit 346 calculates the crack rate of the sliding window 542. Next, the calculation unit 346 slides the left edge of the sliding window 542 200 pixels to the right to set the sliding window 543. Therefore, a region 200 pixels wide from the right end of sliding window 541 overlaps with a region 200 pixels wide from the left end of sliding window 543. Also, a region 400 pixels wide from the right end of sliding window 542 overlaps with a region 400 pixels wide from the left end of sliding window 543. Then, calculation unit 346 calculates the crack rate of sliding window 543.

[0051] In this way, the calculation unit 346 sets the sliding window sequentially by sliding it from the left end to the right end of the top row of the second binary image 54 by an amount that is narrower than the horizontal width of the sliding window, and calculates the crack rate for that sliding window. Next, the calculation unit 346 sets the sliding window by sliding it from the top row of the second binary image 54 by an amount that is narrower than the vertical width of the sliding window. For example, the calculation unit 346 sets the sliding window at the left end of the second row by sliding it 200 pixels down from the sliding window 541, and calculates the crack rate for that sliding window. Thereafter, the calculation unit 346 calculates the crack rate for each sliding window sequentially from the left end to the right end of the second row of the second binary image 54, and then to the right end of the bottom row. Note that the size of the sliding window, the sliding order, the sliding width, etc. are not limited to these.

[0052] Thereafter, the tire inspection support device 300 identifies the maximum value from among the calculated crack rates (S108). Specifically, for example, the identification unit 347 identifies the maximum crack rate from among the crack rates of the sliding window 541, etc.

[0053] Then, the tire inspection support device 300 determines the deterioration level from the identified crack rate (S109). Specifically, the determination unit 348 determines one of multiple level values ​​depending on the crack rate, and sets the determination result as the deterioration level. For example, the determination unit 348 determines the deterioration level as "small" if the crack rate is less than 2.5%, "medium" if the crack rate is 2.5% or more but less than 4.2%, and "large" if the crack rate is 4.2% or more.

[0054] Furthermore, independently of step S109, the tire inspection support device 300 identifies a unit area corresponding to the identified crack rate (S110). That is, the identification unit 347 identifies a sliding window corresponding to the maximum crack rate. Then, the tire inspection support device 300 identifies a position in the captured image that corresponds to the identified unit area (S111). Then, the tire inspection support device 300 generates a display image by adding display information for the identified position to the captured image (S112). For example, the generation unit 344 sets a red frame line surrounding the corresponding area of ​​the identified position in the divided captured image 51 as display information. Then, the generation unit 344 adds the display information to the corresponding area on the divided captured image 51 and generates a display image.

[0055] After steps S109 and S112, the tire inspection support device 300 outputs the maximum crack rate, the determined deterioration level, and the display image to the display device 400 (S113). The tire inspection support device 300 may further output a support message according to the deterioration level. In response to this, the display device 400 displays the maximum crack rate, the determined deterioration level, and the display image received from the tire inspection support device 300 on the screen.

[0056] FIG. 13 is a diagram showing an example of a captured image (display image 55) to which a crack rate, a deterioration level, and display information according to the second embodiment are added. That is, FIG. 13 shows an example in which the crack rate, the deterioration level, and the display information are displayed in association with each other. The display image 55 shows an example in which display information 554 is added to a corresponding area at a specified position within the divided captured image 51 described above. The color of the border of the display information 554 may be red. However, the color is not limited to red and may be any easily distinguishable color. Furthermore, other easily distinguishable display modes, such as thick or dashed lines, may be used instead of colors. The maximum crack rate value 551 indicates that the crack rate in the area of ​​the display information 554 is 3.3%. The deterioration level 552 indicates that the degree of deterioration is determined to be "medium" according to the maximum crack rate value 551. The support message 553 is an example of a message to the mechanic U according to the deterioration level 552. In this example, since the deterioration level 552 is "medium," the support message 553 is "The tire is somewhat deteriorated." The maximum crack rate value 551, the deterioration level 552, the support message 553, and the display information 554 make it easier for the mechanic U to determine that maintenance is necessary for the tire 100. In other words, the tire inspection support device 300 can provide the mechanic U with an objective indicator for determining whether maintenance is necessary for the tire 100, and support the tire inspection.

[0057] In this way, when calculating the index value from the second binary image, the second binary image is scanned while a sliding window of a predetermined size is shifted by a sliding width narrower than the window width. This makes it possible to more accurately determine the position of the maximum crack ratio compared to when the second binary image is simply divided into a grid (for each first unit area) and the index value is calculated. For example, as shown in display information 554 in FIG. 13, it is possible to identify a position that cannot be identified by simply dividing into a grid.

[0058] The first detection unit 343 detects candidate crack regions (first regions) from the captured image of the tire using the first image processing model 312 that has undergone deep learning. However, the candidate crack regions detected by the first image processing model 312 tend to detect the cracks as being thicker than they actually are. Therefore, in this embodiment, the generation unit 344 masks regions other than the first region in the divided captured image 51 to generate a processing target image 53 that leaves a partial image of the first region. The second detection unit 345 then performs binarization processing on the processing target image 53, thereby narrowing down the actual crack regions from the partial images. In other words, in this embodiment, the crack detection process is performed in two stages, making it possible to detect the shape of the crack in detail.

[0059] Furthermore, since the processing target image 53 after masking from the captured tire image has uneven brightness, if the entire image is binarized at once, the improvement in accuracy in narrowing down the cracked area will be limited. Therefore, in this embodiment, the processing target image 53 is divided into multiple unit areas and binarized for each unit area, allowing for detailed analysis and improving detection accuracy. In particular, by adjusting the binarization threshold for each unit area, it is possible to improve detection accuracy by taking uneven brightness into account. Note that the second detection unit 345 may perform second image processing other than binarization processing.

[0060] In this embodiment, the area corresponding to the maximum crack rate is identified and clearly displayed in the captured image. This makes it easier for the mechanic U to visually identify the most deteriorated part of the tire 100 being inspected, making it easier to determine whether maintenance is necessary. This is because a tire is one piece, and if even one large crack is found, the tire needs to be replaced.

[0061] <Other embodiments> In the second embodiment described above, the processing target image 53 is divided into a plurality of unit areas before performing binarization processing and calculating the crack rate, but this is not limiting. For example, the second detection unit 345 may perform binarization processing on the entire processing target image 53 without dividing the processing target image 53, and detect the second area. In this case, the calculation unit 346 may calculate, as an index value, a crack rate based on the ratio between the captured image (divided captured image 51) and the second area (the number of pixels belonging to the second area).

[0062] Furthermore, in the second embodiment described above, display information is added to the unit region corresponding to the maximum crack rate, but this is not limited to this. For example, the identification unit 347 may identify positions in the captured image corresponding to two or more second regions. In this case, the output unit 349 may add multiple pieces of display information to the captured image to distinguishably display each of the multiple identified positions, and output the image. Furthermore, while the above-described embodiments have been described with reference to cracks in tires, the technology disclosed herein can also be applied to cracks in objects and structures other than tires (roads, bridges, buildings, etc.).

[0063] In the above examples, the program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray® disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagated signals.

[0064] The present disclosure is not limited to the above-described embodiments, and may be modified as appropriate without departing from the spirit and scope of the present disclosure. In addition, the present disclosure may be implemented by appropriately combining the respective embodiments.

[0065] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes. (Appendix A1) a first detection means for detecting a first region including a candidate crack portion of the tire by performing a first image processing on a captured image of the tire; a generating means for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection means for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation means for calculating an index value related to cracks in the tire based on the second region; A tire inspection support device comprising: (Appendix A2) The second detection means performs a binarization process on the processing target image as the second image processing. 10. A tire inspection assistance device according to claim A1. (Appendix A3) a determination means for determining a degree of deterioration of the tire based on the index value; a first output means for outputting the degree of deterioration; Further equipped A tire inspection support device according to appendix A1 or A2. (Appendix A4) The device further includes a second output means for outputting the index value. A tire inspection support device according to any one of appendices A1 to A3. (Appendix A5) The generating means extracts the partial image by masking an area other than the first area from the captured image, and generates the processing target image. A tire inspection support device according to any one of appendices A1 to A4. (Appendix A6) the second detection means divides the processing target image into a plurality of first unit areas, and performs the second image processing for each of the first unit areas to detect the second area for each of the first unit areas; The calculation means sets a part of the image to be processed as a second unit area, sets a plurality of second unit areas within the image to be processed by sliding the second unit area in a size narrower than the size of the second unit area, and calculates the index value for each second unit area. A tire inspection support device according to any one of appendices A1 to A5. (Appendix A7) a first specifying means for specifying a second unit area corresponding to one of the index values ​​calculated for each of the second unit areas that satisfies a predetermined condition, and for specifying a position in the captured image that corresponds to the specified second unit area; a third output means for adding display information for identifiably displaying the specified position to the photographed image and outputting the same; Further equipped 1. A tire inspection assistance device as described in Appendix A6. (Appendix A8) The predetermined condition is that the index value is the maximum value among the index values ​​calculated for each of the second unit regions. 1. A tire inspection assistance device as described in Appendix A7. (Appendix A9) The calculation means calculates a crack rate based on a ratio between the second unit area and the second area detected in the second unit area as the index value. A tire inspection support device according to any one of appendices A6 to A8. (Appendix A10) The calculation means calculates a crack rate based on a ratio between the captured image and the second region as the index value. A tire inspection support device according to any one of appendices A1 to A5. (Appendix A11) a second specifying means for specifying a position in the captured image corresponding to the second region; a fourth output means for adding display information for identifiably displaying the specified position to the photographed image and outputting the photographed image; Further equipped A tire inspection support device according to any one of appendices A1 to A5 and A10. (Appendix B1) The computer Detecting a first region including a candidate crack portion of the tire by performing a first image processing on the captured image of the tire; generating a processing target image by extracting a partial image corresponding to the first region from the captured image; detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; Calculating an index value related to cracks in the tire based on the second region. A method for assisting in tire inspection. (Appendix C1) a first detection process for detecting a first region including a candidate crack portion of the tire by performing a first image process on a captured image of the tire; a generation process for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection process for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation process for calculating an index value related to cracks in the tire based on the second region; A non-transitory computer-readable medium storing a tire inspection support program that causes a computer to execute the above.

[0066] Although the present invention has been described above with reference to the embodiments (and examples), the present invention is not limited to the above-described embodiments (and examples). Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Explanation of symbols]

[0067] 1. Tire inspection support device 11 First detection unit 12 Generation part 13 Second detection unit 14 Calculation section 1000 Tire Inspection Support System 100 tires 200 cameras 300 Tire inspection support device 310 Storage section 311 Tire Inspection Assistance Program 312 First Image Processing Model 320 memory 330 IF Section 340 Control Unit 341 Acquisition Department 342 Pretreatment section 343 First detection unit 344 Generation part 345 Second detection unit 346 Calculation Unit 347 Specific part 348 Judgment section 349 Output Section 400 display device U Mechanic 51 split shot images 52 Detection Results 53 Processing target image 531 First Unit Area 532 First Unit Area 533 First Unit Area 54 Second binary image 541 Sliding Window 542 Sliding Window 543 Sliding Window 55 Display Images 551 Maximum crack rate 552 Deterioration degree 553 Messages of Support 554 Display information

Claims

1. a first detection means for detecting a first region including a candidate crack portion of the tire by performing a first image processing on a photographed image of the tire; a generating means for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection means for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation means for calculating an index value related to cracks in the tire based on the second region; Equipped with the second detection means divides the processing target image into a plurality of first unit areas, and performs the second image processing for each of the first unit areas to detect the second area for each of the first unit areas; the calculation means sets a part of the image to be processed as a second unit area, sets a plurality of second unit areas within the image to be processed by sliding the second unit area to a size narrower than the size of the second unit area, and calculates the index value for each second unit area; Tire inspection support device.

2. The second detection means performs a binarization process on the processing target image as the second image processing. The tire inspection support device according to claim 1 .

3. a determination means for determining a degree of deterioration of the tire based on the index value; a first output means for outputting the degree of deterioration; Further equipped The tire inspection support device according to claim 1 or 2.

4. The device further includes a second output means for outputting the index value. The tire inspection support device according to any one of claims 1 to 3.

5. The generating means extracts the partial image by masking an area other than the first area from the captured image, and generates the processing target image. The tire inspection support device according to any one of claims 1 to 4.

6. a first specifying means for specifying a second unit area corresponding to one of the index values ​​calculated for each of the second unit areas that satisfies a predetermined condition, and for specifying a position in the captured image that corresponds to the specified second unit area; a third output means for adding display information for identifiably displaying the specified position to the photographed image and outputting the photographed image; Further equipped The tire inspection support device according to any one of claims 1 to 5.

7. The calculation means calculates a crack rate based on a ratio between the captured image and the second region as the index value. The tire inspection support device according to any one of claims 1 to 5.

8. The computer detecting a first region including a candidate crack portion of the tire by performing a first image processing on the captured image of the tire; generating a processing target image by extracting a partial image corresponding to the first region from the captured image; detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; Calculating an index value related to cracks in the tire based on the second region; When detecting the second region, the processing target image is divided into a plurality of first unit regions, and the second image processing is performed for each of the first unit regions, thereby detecting the second region for each of the first unit regions; When calculating the index value, a part of the image to be processed is set as a second unit area, and a plurality of second unit areas are set within the image to be processed by sliding the second unit area to a size narrower than the size of the second unit area, and the index value is calculated for each second unit area. A method for assisting in tire inspection.

9. a first detection process for detecting a first region including a candidate crack portion of the tire by performing a first image process on a captured image of the tire; a generation process for generating a processing target image by extracting a partial image corresponding to the first region from the captured image; a second detection process for detecting a second region indicating the crack portion from the first region by performing second image processing on the processing target image; a calculation process for calculating an index value related to cracks in the tire based on the second region; on the computer, In the second detection process, the processing target image is divided into a plurality of first unit areas, and the second image process is performed for each of the first unit areas, thereby detecting the second area for each of the first unit areas; In the calculation process, a part of the image to be processed is set as a second unit area, and a plurality of second unit areas are set within the image to be processed by sliding the second unit area to a size narrower than the size of the second unit area, and the index value is calculated for each second unit area. Tire inspection assistance program.

Images