Tire damage evaluation device, tire damage evaluation method, and tire damage evaluation program

Abstract

The tire damage evaluation apparatus includes a processor configured to acquire a tread image obtained by photographing a tread portion of a tire and quantify a damage state of a surface area of the tread portion based on the tread image.

Description

Technical Field

[0001] The present disclosure relates to a tire damage assessment device, a tire damage assessment method, and a tire damage assessment program. Background Art

[0002] Conventionally, inspections of the appearance and damage condition of tires have been performed by human visual inspection. Additionally, in the appearance inspection of tires, for example, an assessment of the surface condition of the tire based on an image of the tire has been employed (see, for example, Patent Document 1). Prior Art Documents Patent Documents

[0003] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2021-42988 Summary of the Invention Technical Problem to be Solved by the Invention

[0004] However, the inspection method of visually inspecting the appearance of tires by humans is greatly influenced by experience and professional skills, lacks objectivity, and remains a qualitative assessment. In addition, conventional image analysis techniques only detect whether a tire is a qualified product or a non-qualified product on a production line, and cannot quantify and evaluate the surface condition of the tread surface of a tire after the tire has been put into use.

[0005] Therefore, in view of these points, an object of the present disclosure is to provide a tire damage assessment device, a tire damage assessment method, and a tire damage assessment program that can quantify and evaluate the surface condition of the tread surface of a tire. Solution to the Problem

[0006] (1) A tire damage assessment device according to an aspect of the present disclosure includes a processor configured to acquire a tread image obtained by photographing a tread surface of a tire and quantify a damage state of a surface area of the tread surface based on the tread image.

[0007] (2) A tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (1) above, wherein the tread surface includes a plurality of layers arranged radially along the tire, and the processor is preferably configured to generate a segmented image in which a surface area of the tread surface included in the tread image is segmented by a layer exposed in the surface area.

[0008] (3) A tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (2) above, wherein the plurality of layers preferably include an outer rubber layer constituting the tread surface of the tread, a metal cord layer arranged radially inward along the tire from the outer rubber layer, and an inner rubber layer located between the outer rubber layer and the metal cord layer.

[0009] (4) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (2) or (3) above, wherein the processor is preferably configured to perform processing to: input a tread image of the tire into a first trained model, the first trained model being trained using first training data having tread images of a plurality of sample tires as input and having a segmented image obtained by segmenting the surface region of the sample tire included in the tread image using the layer exposed on the surface as output; and output the segmented image of the tire.

[0010] (5) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in any one of (2) to (4) above, wherein the processor is preferably configured to divide a segmented image of the tire into a plurality of strip regions extending along the width direction of the tire, align the positions of the plurality of strip regions, make the tread portion in the strip regions consistent in the width direction, and generate a histogram obtained by accumulating the pixels of the plurality of strip regions for each position in the width direction at least one layer exposed in the surface region.

[0011] (6) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device according to any one of (1) to (5) above, wherein the processor is preferably configured to extract at least one of the following from the tread image: a groove portion for indicating a groove of the tread surface, a local damage portion for indicating local damage to a surface area of ​​the tread surface, and a repair portion for indicating a location of previously occurring repair damage in the tire.

[0012] (7) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device according to any one of (1) to (6) above, wherein the processor is preferably configured to perform processing to: input a tread image of the tire into a second trained model, the second trained model being trained using second training data having tread images of a plurality of sample tires as input and having an image of a locally damaged portion indicating local damage to the tread portion of the sample tire included in the tread image as output; and output an image of the locally damaged portion of the tire.

[0013] (8) A tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (7) above, wherein the processor is preferably configured to calculate the ratio of the area of ​​the region of the local damage to the area of ​​the surface region of the tread.

[0014] (9) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (7) or (8) above, wherein the processor is preferably configured to calculate the centroid position of the region of the local damage.

[0015] (10) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in any one of (2) to (5) above, wherein the processor can quantify the damage state of the surface region of the tire in the outermost exposed portion of the tire among the plurality of layers based on the tread image.

[0016] (11) A tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (10) above, wherein the processor is preferably configured to extract an image of a region including the surface area of ​​the tread from the tread image, divide the image into images of a plurality of sub-regions, and quantify the damage state of the surface area for each sub-region image.

[0017] (12) A tire damage assessment device according to an embodiment of the present disclosure is a tire damage assessment device described in (10) or (11) above, wherein the processor is preferably configured to perform processing to: input a tread image of the tire into a third trained model, the third trained model being trained using third training data having tread images of a plurality of sample tires as input and having information as output for quantifying the damage state of the surface region in the outermost exposed portion of the tire among the plurality of layers included in the tread image; and output information for quantifying the damage state of the surface region.

[0018] (13) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in (12) above, wherein the damage state of the surface region in the third training data is preferably quantified based on at least one of the number of scratches included in the surface region and the depth of the scratches.

[0019] (14) The tire damage assessment device according to an embodiment of the present disclosure is the tire damage assessment device described in any one of (1) to (13) above, wherein the damage state includes the state of defects and scratches on the tire, and the processor is preferably configured to estimate the wear characteristics of the tire based on information for quantifying the state of the defects and scratches.

[0020] (15) A tire damage assessment device according to an embodiment of the present disclosure is a tire damage assessment device described in any one of (1) to (14) above, preferably including at least one of a display unit and a communication unit, the display unit being configured to display information for quantifying the damage state of the surface area of ​​the tread, and the communication unit being configured to send the information to an external device.

[0021] (16) A tire damage assessment method according to one aspect of the present disclosure is a tire damage assessment method performed by an information processing device, comprising: acquiring a tread image obtained by photographing the tread portion of the tire; and quantifying the damage state of a surface area of ​​the tread portion based on the tread image.

[0022] (17) A tire damage assessment procedure according to one aspect of the present disclosure is configured such that a computer performs: processing to obtain a tread image obtained by photographing the tread of the tire; and processing to quantify the damage state of the surface area of ​​the tread based on the tread image. The effects of the invention

[0023] According to this disclosure, the damage status of the surface area of ​​the tread of a tire can be quantified and assessed. Attached Figure Description

[0024] In the attached diagram: Figure 1 This is a block diagram illustrating a schematic configuration of a tire damage assessment system including a tire damage assessment device according to an embodiment of the present disclosure; Figure 2 The description should be made by Figure 1 A cross-sectional view of an example tire evaluated by the tire damage assessment system; Figure 3 This is a flowchart illustrating a method for generating a first trained model that segments the surface region of the tread based on images of multiple sample tires; Figure 4 This is an example image depicting the tread of a sample tire; Figure 5 This is a diagram depicting an example of a segmented image generated from an image of the fetal face; Figure 6 This is a diagram illustrating a method for generating histograms from segmented images; Figure 7 This is a diagram depicting examples of the grooves, damaged areas, and repaired areas of a tire; Figure 8 This is a flowchart illustrating a method for generating a third trained model that quantifies the damage level of the surface region of the tread based on images of multiple sample tires. Figure 9 This is an example diagram depicting a tire tread image captured using a camera unit; Figure 10 It is a description of the execution used to determine Figure 9 An example diagram of the processing of rectangular regions in a tire tread image; Figure 11 It is a description Figure 10 An example image showing how the damaged area of ​​a rectangular region in an image is masked and divided into sub-regions; Figure 12 It is a depiction of Figure 11 A graph of the image in which sub-regions of the tread image are assigned numerical values ​​to quantify the damage status; Figure 13 This is a flowchart illustrating a tire damage assessment method according to embodiments of the present disclosure; and Figure 14 This is a block diagram illustrating a schematic configuration of another tire damage assessment system according to another embodiment of the present disclosure. Detailed Implementation

[0025] Tires used on vehicles operating in special environments such as mines are susceptible to damage (including defects and scratches on the tread surface) due to travel on roads other than paved roads. Even minor tire damage raises concerns that it might serve as the starting point for rapid tire wear progression. By quantitatively understanding the relationship between tire damage and tire wear progression, the relationship between tire damage and tread pattern can be clarified, and this relationship can be utilized in tire design. Furthermore, it is anticipated that information can be provided to tire users by estimating the appropriate timing for tire rotation or replacement based on the tire's damage condition.

[0026] In this application, "damage" means that the normal state is at least partially impaired. "Damage" to the tire surface includes wear, defects, and scratches on the tire surface. Wear on the tire surface refers to the state where the tire surface is eroded or scraped off due to friction with the ground. Defects on the tire surface refer to the state where a portion of the tire surface is missing. Scratches on the tire surface refer to the state where abrasion or cuts have occurred on the tire. In this application, "localized damage" refers to defects, scratches, and marks occurring in a relatively narrow area, and can be identified by an operator in terms of location and extent based on an image of the tire surface.

[0027] In the following description, embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0028] (Tire Damage Assessment System) like Figure 1As illustrated, the tire damage assessment system 1 according to an embodiment of the present disclosure includes a tire damage assessment device 10 and a camera device 20. Figure 2 The diagram depicts a cross-sectional view of an example tire 30 to be evaluated by the tire damage assessment system 1. In the tire damage assessment system 1, the tire damage assessment device 10 analyzes and quantifies the damage state of the surface area of ​​the tread portion 35 of the tire 30 based on images of the tread portion 35 of the tire 30 captured by the imaging device 20. The tread portion 35 refers to the part of the tire 30 that is in direct contact with the road. Hereinafter, images of the tread portion 35 will be referred to as tread images.

[0029] The tire damage assessment device 10 is an information processing device, such as a server computer, workstation, or PC (personal computer). The tire damage assessment device 10 can be installed at the user's site or can be remotely located via a network. Furthermore, the components and functions of the tire damage assessment device 10 described below can be distributed across multiple different hardware devices. Additionally, the tire damage assessment device 10 can be a small information processing device, such as a portable information terminal or smartphone. In the embodiment described below, the tire damage assessment device 10 is described as a fixed computer. The tire damage assessment device 10 acquires tire tread images from the camera device 20 and performs processing including image processing.

[0030] The camera device 20 is equipped with an optical system and imaging elements. The imaging elements include a CCD image sensor (charge-coupled device image sensor) and a CMOS image sensor (complementary MOS image sensor). The imaging elements convert the image formed by the optical system on its light-receiving surface into an electrical signal. In the following embodiments, the camera device 20 includes a set of optical systems and imaging elements, and captures images of the tread surface 35 of the tire 30 from one direction while the tire 30 is removed from the vehicle and laid flat. However, the camera device 20 can be configured to include multiple sets of optical systems and imaging elements, and capture images of the tread surface 35 from multiple directions. The camera device 20 can be configured to capture images of the tire 30 from all directions (360 degrees) along the circumference of the tire 30. The camera device 20 can be a digital camera or a portable information terminal equipped with a camera. The camera device 20 can also be a device specifically designed for the tire damage assessment system 1.

[0031] (Configuration of tire damage assessment equipment) In one embodiment, the tire damage assessment device 10 includes a communication unit 11, an input unit 12, an output unit 13, a processor 14, and a storage unit 15.

[0032] Communication unit 11 provides communication functionality for communicating with external devices of tire damage assessment equipment 10. Communication unit 11 includes a communication module for connecting to a communication network such as the Internet. Communication unit 11 may include a communication module compatible with mobile communication standards such as 4G (4th generation) or 5G (5th generation). Communication unit 11 may include a communication module compatible with wired LAN standards (e.g., 1000BASE-T). Communication unit 11 may include a communication module compatible with wireless LAN standards (e.g., IEEE 802.11). Communication unit 11 may include a communication module compatible with short-range communication such as Bluetooth and BLE.

[0033] The communication unit 11 can be configured to connect to the camera device 20 via wired or wireless communication and receive tire tread images captured by the camera device 20. The communication unit 11 can be configured to enable the transmission and reception of information with the information processing device for machine learning, described later, via a network such as the Internet.

[0034] Input unit 12 includes one or more input interfaces for tire damage assessment device 10. Input interfaces may include, for example, physical keys, capacitive keys, pointing devices, touchscreens integrated with the display, microphones, and cameras. Input unit 12 includes an input interface for retrieving information from portable storage media such as memory cards. Tire damage assessment device 10 can acquire tire tread images captured by camera device 20 via the storage media.

[0035] Output unit 13 includes one or more output interfaces. Output interfaces may include, for example, a display, a printer, and a speaker. As a display, for example, an LCD (liquid crystal display) or an OLED (organic EL) display can be used. That is, output unit 13 includes a display unit. Input unit 12 may include an output interface for writing information to a portable storage medium such as a memory card. Output unit 13 can output information obtained through the operation of tire damage assessment device 10 to the user.

[0036] Processor 14 includes one or more processors. The processor can be a general-purpose processor, such as a CPU (Central Processing Unit) that executes a program defining control procedures, or a dedicated processor for specific processing. Processor 14 performs computational processing to execute the various functions of the tire damage assessment device 10 and controls the various components of the tire damage assessment device 10.

[0037] The processor 14 can perform various image processing operations on the tire tread image acquired from the camera device 20. The processor 14 can use at least one of the first to third trained models stored in the storage unit 15, which will be described later, to evaluate the damage state of the surface region of the tire tread 35.

[0038] Storage unit 15 includes one or more memories. The memories may be, for example, semiconductor memories, magnetic memories, or optical memories, but are not limited to these, and can be any type of memory. Semiconductor memories may include volatile memories such as RAM (Random Access Memory) and non-volatile memories such as ROM (Read Only Memory) and flash memory. Magnetic memories may include, for example, hard disks and magnetic tapes. Optical memories may include, for example, CDs (Compact Discs), DVDs (Digital Versatile Discs), and BDs (Blu-ray Discs).

[0039] Storage unit 15 stores information required for processing by the processor 14 of the tire damage assessment device 10, as well as information generated as a result of processing by the processor 14. For example, storage unit 15 stores tread images captured by the camera device 20, images processed from the tread images, and information related to the results of assessing the damage status of the surface areas of the tire 30 based on the tread images. Storage unit 15 may store the tire damage assessment program executed by the processor 14 of the tire damage assessment device 10, as well as first to third trained models. The tire damage assessment program may be provided in the form of a non-transitory storage medium such as a CD, DVD, or USB (Universal Serial Bus) memory.

[0040] (Example tire configuration to be evaluated) refer to Figure 2 The cross-sectional view shown below illustrates an example of a tire 30 to be evaluated by the tire damage assessment device 10. In one embodiment, the tire 30 is a tire mounted on vehicles used in special environments such as mines, and on special vehicles such as construction and civil engineering vehicles. Due to use in harsh environments, the tire 30 is susceptible to various types of damage. Furthermore, the tire 30 is often used while suffering damage until the end of its service life.

[0041] Figure 2The illustrated tire 30 includes a pair of bead portions 31, a pair of sidewall portions 32, and a tread portion 35. The tread portion 35 has an outer rubber layer 33 and an inner rubber layer 34 sequentially arranged from the radially outer side of the tire, both of which are connected to the sidewall portions 32. The tire 30 also includes a carcass 36 and a metal cord layer 37. The carcass 36 extends circumferentially between the pair of bead portions 31 and reinforces the bead portions 31, sidewall portions 32, and tread portion 35. The metal cord layer 37 is disposed radially outer of the crown portion of the carcass 36 and radially inner of the inner rubber layer 34. In other words, the outer rubber layer 33, inner rubber layer 34, and metal cord layer 37 are arranged sequentially from the radially outer side of the tire 30.

[0042] here, Figure 2 The outer rubber 33 of the illustrated tire 30 is generally referred to as the "crown rubber". The inner rubber 34 is generally referred to as the "base rubber". The metal cord layer 37 is generally referred to as the "belt".

[0043] Figure 2 The tire 30 illustrated has a carcass 36 composed of a single carcass ply, which is formed by coating a plurality of parallel cords with coated rubber. The carcass 36 includes a main body extending annularly between bead cores 38 embedded in the bead portion 31, and a folded portion that is radially wound around each bead core 38 in the tire width direction from the inside to the outside. The number of plies and the structure of the carcass 36 are not limited thereto.

[0044] also, Figure 2 The illustrated tire 30 has a metal cord layer (belt) 37 consisting of two belt layers. The number of belt layers constituting the metal cord layer 37 is not limited to this, and the number of belt layers can be three or more. Here, the belt layer is typically a rubber coating layer of metal cords (preferably steel cords) extending obliquely relative to the tire equator. The two belt layers are laminated such that the metal cords constituting the belt layer cross each other across the tire equator, thereby forming the metal cord layer 37.

[0045] The tire 30 includes an outer rubber layer 33 forming the tread surface of the tread 35, a metal cord layer 37 disposed radially inside the outer rubber layer 33, and an inner rubber layer 34 located between the outer rubber layer 33 and the metal cord layer 37. A tread pattern is formed on the surface of the outer rubber layer 33. The rubber layers of the tread 35 of the tire 30 are not limited to a two-layer structure of outer rubber layer 33 and inner rubber layer 34, but can be a single-layer structure or a structure having three or more layers. The tire damage assessment device 10 of this disclosure can be suitably applied to the assessment of damage in a tire 30 having such a tread 35 configuration.

[0046] (A method for dividing the surface area of ​​a tire) The processor 14 of the tire damage assessment device 10 acquires a tread image, including an image of the tread portion 35 of the tire 30 captured by the camera device 20, via the communication unit 11 or the input unit 12. The processor 14 generates a segmented image in which the surface area of ​​the tread portion 35 included in the acquired tread image is segmented using the layers exposed in the surface area. The processor 14 can segment the surface area into multiple regions (see...). Figure 4 and Figure 5 These multiple areas include a first area 41 covered by the outer rubber layer 33, a second area 42 exposed by the inner rubber layer 34, and a third area 43 exposed by the metal cord layer 37.

[0047] Various image processing methods can be used to generate segmented images. In one embodiment, machine learning can be used to generate segmented images. Processor 14 can be configured to input a tread image into a first trained model that has been trained by an information processing device for machine learning, and output a segmented image of tire 30. The information processing device for machine learning can be the same computer as the tire damage assessment device 10 or a different computer. In the latter case, the information processing device for machine learning can be configured to communicate with the tire damage assessment device 10.

[0048] Based on Figure 3 The flowchart in the document describes the method used to generate the first trained model.

[0049] First, the operator takes images of the treads of multiple tires (step S101). These multiple tires are referred to as sample tires. Sample tires can include tires still in use and tires that have reached the end of their service life. A large number of images of sample tires under different conditions are taken and accumulated.

[0050] The operator segments the surface area 35 of the tread portion in the tread image of each sample tire using the exposed layers in the surface area on the screen of the working computer (step S102). The operator can be a skilled technician who can determine which of the outer rubber 33, inner rubber 34, and metal cord layer 37 is exposed in a given area by viewing the image of the sample tire.

[0051] Here, the fact that all layers of tire 30 except the outermost layer are exposed on the tire surface means that the upper layers have been worn away and lost due to vehicle movement. Therefore, areas where, for example, the inner rubber layer 34 or the metal cord layer 37 is exposed are included in the damaged area. It can be considered that the area where the metal cord layer 37 is exposed is more severely damaged than the area where the inner rubber layer 34 is exposed.

[0052] Figure 4An example is shown of a sample tire tread image 40 and an example of the operator's segmentation of the surface region on the tread image 40. (See example...) Figure 4 As illustrated, the operator uses the GUI (Graphical User Interface) of the working computer to sequentially specify the boundaries of different exposed layers on an image of the tread layer 35 of the sample tire. These boundaries include, for example, the boundary between a first region 41 of the outermost rubber layer 33 and a second region 42 of the inner rubber layer 34, and the boundary between the second region 42 and a third region 43 of the metal cord layer 37. By specifying boundaries to enclose the damaged areas (e.g., the second region 42 and the third region 43), the operator generates a segmented image of the surface regions of the tread image 40. Thus, the first region 41, the second region 42, and the third region 43 are segmented on the tread image 40 of the sample tire. The working computer can be any computer and can be the same computer as the tire damage assessment device 10 and / or the information processing device used for machine learning.

[0053] Next, the first training data for machine learning is generated using the tread image 40 of the sample tires as input and the data of the segmented image of the surface region of the tread image 40 being segmented as output (step S103). The first training data includes a large amount of data generated from a large number of sample tires.

[0054] Using the first training data generated in step S103, the information processing device for machine learning performs machine learning (step S104). The information processing device for machine learning is equipped with a function approximator capable of learning the input-output relationship and learns the first training data to generate a first trained model (step S105). The first trained model is also referred to as a segmentation model. The function approximator can perform machine learning using known machine learning algorithms such as neural networks or deep learning. When the first trained model is input along with the tread image 40 of the tire 30, it outputs a segmented image of the surface region of the tread portion 35 of the tire 30. By learning data from more tread images 40 from sample tires, the accuracy of machine learning can be improved, and the accuracy of segmentation of the surface region of the tire can be enhanced.

[0055] according to Figure 3 The first trained model generated by the flowchart in the diagram is stored in the storage unit 15 of the tire damage assessment device 10. The processor 14 inputs the tread image 40 of the tire 30 captured by the camera device 20 into the first trained model read from the storage unit 15, and outputs a segmented image of the surface area of ​​the tread portion 35 of the tire 30, thereby generating a segmented image. The segmented image can be displayed on a display, which serves as the output unit 13. As a result, the damage state caused by wear on the tread portion 35 of the tire 30 is visualized.

[0056] Figure 5 An example of a segmented image generated from an image of the tread region 35, showing the division of the surface area. Figure 5 The image depicts a first region 41 (where the outer rubber 33 is exposed on the surface), a second region 42 where the inner rubber 34 is exposed, and a third region 43 where the metal cord layer 37 is exposed. Each region can be displayed in a different color. From the area of ​​each region, the user can visually and to some extent quantitatively assess the degree of damage to the surface area of ​​the tire 30. The processor 14 can quantify the area ratio (which is the ratio of the area of ​​each of the first region 41, the second region 42, and the third region 43 to the entire surface area of ​​the tread 35) into a ratio of pixels in the image.

[0057] Furthermore, processor 14 can process segmented images to identify the position of the inner rubber 34 and the metal cord layer 37, which are mostly exposed in the width direction of the tire 30. For this purpose, as... Figure 6 As illustrated, the processor 14 divides the segmented image of the tire 30 into multiple strip-shaped regions 45, which are narrow in the radial direction of the tire 30 and extend long in the width direction of the tire 30. The term "strip" can also be referred to as "band" or "strip".

[0058] The processor 14 aligns the positions of the multiple strip regions 45 so that the positions of the tread portion 35 of the tire 30 in the strip regions 45 are consistent in the width direction. Figure 6 The strip region 45 is depicted before alignment. Depending on the camera angle, the tread image 40 of the tire 30 can be offset from the outer edge of the image center. Furthermore, the width of the tire 30 can appear narrower outside the image center. The processor 14 aligns the center position of the strip region 45 and makes the strip region 45 aligned in the width direction of the tire 30 (that is,...). Figure 6 The lengths in the vertical direction are consistent. Through this operation, the positions of the tires 30 in the width direction of each strip region 45 are consistent. This operation is called normalization.

[0059] Next, the processor 14 generates a histogram on the image formed by arranging the normalized strip regions 45. In this histogram, for each position in the width direction of the tire 30, pixels corresponding to one or more of the first region 41, the second region 42, and the third region 43 are accumulated. Figure 6 In the example, histograms of the second region 42 and the third region 43 are presented on the right. This allows the user to determine the location and size of most of the exposed inner rubber layer 34 and metal cord layer 37 in the width direction of the tire 30. In other words, the user can determine the location and size of areas on the surface of the tread 35 that have been extensively damaged due to wear, etc.

[0060] (Extraction of localized damage on the tire surface) In use Figures 3 to 6 In the above description, it is assumed that the tire damage assessment device 10 generates a segmented image in which the surface area of ​​the tire 30 is divided into areas where the outer rubber layer 33, the inner rubber layer 34, and the metal cord layer 37 are exposed. In addition to these, such as Figure 7 As illustrated, the tire damage assessment device 10 can be configured to extract a groove 51 for indicating the tread grooves, a local damage section 52 including local damage such as defects, scratches, and cuts, and a repair section 53 for indicating previously repaired areas. Figure 7 In the middle, the local damage portion 52 included in the small area 54 is magnified and displayed.

[0061] To extract the groove 51, the locally damaged portion 52, and the repaired portion 53, it is possible to use... Figure 3 The flowchart describes a machine learning-like trained model generated by machine learning. In particular, even small localized damage 52 of tire 30 affects the wear progression of tire 30.

[0062] The process for extracting the localized damaged area 52 will be described. For example... Figure 3 As shown in the flowchart, the operator extracts localized damage 52 from tread images 40 of multiple sample tires and records the region on the tread image 40. For example, the operator performs defect filling on the tread image 40. This generates second training data, using the tread image 40 as input and the region of the localized damage 52 on the tread image 40 as output. A second trained model can be generated by performing machine learning on an information processing device for machine learning using the second training data. The tire damage assessment device 10 can store the second trained model in the storage unit 15. The processor 14 can perform a process that inputs the tread image 40 of the tire 30 into the second trained model read from the storage unit 15 and outputs the localized damage 52 of the tire 30. By increasing the number of sample tires used to generate the second training data, the accuracy of extracting the localized damage 52 can be improved.

[0063] When the processor 14 extracts the local damaged portion 52, the processor 14 can calculate the centroid position (G) of each local damaged portion 52. x G y In this way, processor 14 can present the location and area of ​​the localized damage 52 to the user. Processor 14 can also calculate the area ratio, which is the ratio of the total area of ​​the localized damage 52 to the area of ​​the entire surface area of ​​the tread 35. Therefore, processor 14 can quantify the size of the localized damage 52.

[0064] In the above description, it is assumed that the second trained model outputs information related to the localized damage portion 52 by inputting the tread image 40. In addition to or instead of the localized damage portion 52, the second trained model can be configured to output the groove portion 51 and / or the repair portion 53. The information related to the groove portion 51 and the repair portion 53 can be appropriately used to assess the tire's damage condition.

[0065] The first and second trained models can be generated as a single integrated trained model instead of as separate trained models. By inputting the tread image 40 into the tire damage assessment device 10, the processor 14 can be configured to generate a segmented image and extract minute localized damage portions 52 present in the first region 41 of the segmented image. By continuously monitoring the localized damage portions 52, the user can monitor the process by which these minute localized damage portions 52 lead to significant wear of the tire 30. The user can observe the relationship between the shape of the tread pattern 35 and the progression of tire wear on the tire 30, and utilize the observations in the design of the tire tread pattern.

[0066] (Detection of minor damage level) Even in areas where the inner rubber 34 or metal cord layer 37 is not exposed on the surface of tire 30 (that is, in areas of the outer rubber 33), there may be numerous fine scratches that are difficult to identify individually. In one embodiment, the tire damage assessment device 10 can quantify relatively small damage conditions, such as scratches in the tread 35 of tire 30. The damage condition of the surface area of ​​tire 30 can be determined by analyzing the tread image 40 via image processing. In one embodiment, machine learning can be used in the determination of the damage condition.

[0067] Figure 8 This is a flowchart illustrating a method for generating a third trained model for assessing the damage state of tire 30 using machine learning based on tread image 40. Figure 8 The processing in the flowchart can be performed by an information processing device for machine learning. This information processing device for machine learning can be the same computer as or a different computer from the information processing device used to generate the first trained model and / or the second trained model.

[0068] First, the information processing device for machine learning acquires tread images 40 of multiple sample tires taken by the operator (step S201). Figure 9 This is a diagram depicting the tread image 40 of the sample tire thus acquired. It is expected that tread images of multiple sample tires will be captured at similar angles and magnifications to include the entire tire.

[0069] Next, a rectangular region 55 including the tread portion 35 is extracted from the tread image 40 of the sample tire (step S202). The rectangular region 55 can be extracted as the smallest rectangle including the tread portion 35 in the image of the sample tire. In one embodiment, to identify the tread portion 35 of the sample tire, the tread image 40 can be input into a first trained model to output a segmentation image. By using the first trained model, such as... Figure 10 As illustrated, regions including a first region 41, a second region 42, and a third region 43 are extracted from the tread image 40 of the sample tire. Figure 10 The tire tread area 35 (not shown in the image). Figure 10 In the example, the surface area of ​​the sample tire is mainly the first region 41 where the outer rubber 33 is exposed, and partly the second region 42 where the inner rubber 34 is exposed. The rectangular region 55 is defined as including the first region 41.

[0070] Next, the information processing device for machine learning divides the extracted rectangular region 55 into n×m sub-regions 57 in both the vertical and horizontal directions. For example, in Figure 11 In the example, the segmented region is divided into eight sub-regions 57 arranged in a 2×4 pattern. n and m can be set to any natural numbers. In one embodiment, the information processing device for machine learning can apply a mask 56 to exclude regions with large damage within the rectangular region 55 from subsequent processing. Regions with large damage include, for example, second regions 42 and third regions 43, which were identified as areas where the inner rubber layer 34 or metal cord layer 37 was exposed during analysis using a first trained model. In other embodiments, regions with large damage may include regions where localized damage 52 was extracted during analysis using a second trained model.

[0071] Thus, for each sub-region 57 after division, the operator determines the damage level according to predetermined criteria and quantifies the damage level (step S204). For example... Figure 12 As illustrated, damage levels can be determined in five stages from 0 to 4. For example, damage level 0 indicates a very small scratch, and damage level 4 indicates a significant scratch. Predetermined criteria may include the number of scratches within sub-region 57 or per unit area, as well as criteria related to scratch depth. Because there are many tiny scratches in the surface area of ​​the tread 35 of the sample tire, unlike the methods used to generate the first and second trained models, the operator does not extract individual scratches. The information processing device for machine learning acquires data quantified by the damage levels determined by the operator.

[0072] A third training data is generated by creating and accumulating a large dataset (in which the tread image 40 is used as input and the quantified data of the damage level of each sub-region 57 of the tread image 40 is used as output). (Step S205)

[0073] The information processing device for machine learning uses the third training data generated in step S205 to perform machine learning (step S206). The information processing device for machine learning includes a function approximator capable of learning input-output relationships and generates a third trained model by learning from the third training data (step S207). The third trained model can also be referred to as a scratch prediction model. When the third trained model is input with the tread image 40 of the tire 30, the third trained model outputs numerical values ​​indicating the damage state of each sub-region 57 segmented in the surface region of the tread area 35 of the tire 30. This allows for an objective assessment of the damage state of the surface region, such as scratches that are difficult to determine through visual inspection of the surface region of the tire 30. By increasing the number of sample tires used to generate the third training data, the accuracy of assessing local conditions can be improved.

[0074] according to Figure 8 The third trained model generated by the flowchart is stored in the storage unit 15 of the tire damage assessment device 10. The processor 14 can input the tread image 40 of the tire 30 captured by the camera device 20 into the third trained model and output information for quantifying the damage state of each sub-region 57 divided in the surface region of the tread 35. In this way, the tire damage assessment device 10 can assess the damage state of the surface region based on the position of the sub-regions 57 after the surface region of the tread 35 is divided. The processor 14 can perform this processing on the first region 41, which is the outermost layer of the tread 35 of the tire 30. Note that the information for quantifying the damage state output by the third trained model can be output as a more refined value than the damage level determined by the operator in step S204. For example, if the operator determines the damage state as a natural number from 0 to 4 in step S204, the information for quantifying the damage state output by the third trained model can include a decimal between 0 and 4. Processor 14 can convert information used to quantify damage status, including decimals, into integer values ​​(e.g., from 1 to 100).

[0075] (Tire damage assessment and treatment) Next, refer to Figure 13 The flowchart illustrates an example of a series of processes in a tire damage assessment method executed by the processor 14 of the tire damage assessment device 10, which has already acquired images of the captured tire 30. The tire damage assessment device 10 can perform... Figure 13All processes in the flowchart, or only a portion of these processes can be executed.

[0076] First, the processor 14 acquires a tread image 40 obtained by photographing the tread portion 35 of the tire 30 via the communication unit 11 or the input unit 12. The processor 14 then performs preprocessing on the tread image 40 as required for the following processing (step S301). The preprocessing includes processing to normalize the size of the acquired image. For this purpose, the processor 14 may compress or expand the image during preprocessing.

[0077] Next, the processor 14 inputs the tread image 40 into the first trained model and the second trained model, and segments the tread image 40 into a segmented image including the first region 41 to the third region 43 (step S302). The segmentation in step S302 includes extracting the local damage portion 52. Thus, the processor 14 can generate an image as shown below. Figure 5 The illustrated segmented image, and extracts such as Figure 7 The illustrated local damage section 52. The processor 14 can display each region of the segmented image in a different color on a display unit included in the output unit 13. The processor 14 stores information related to each region and information related to the local damage section 52 in the storage unit 15.

[0078] Processor 14 performs post-processing on the image of the second region 42, which is the area where the inner rubber layer 34 is exposed (step S303). For example, processor 14 can generate an image as shown in the reference image. Figure 6 The histogram used to indicate the distribution of pixels in the second region 42 along the width direction of the tire 30. Furthermore, for example, the processor 14 can calculate, for example, the size and / or area ratio of the surface area of ​​the second region 42 occupying the tread portion 35 of the tire 30. The processor 14 outputs the position, size, and / or area ratio of the second region 42 in the surface area of ​​the tread portion 35 of the tire 30 via the output unit 13, or stores them in the storage unit 15.

[0079] Processor 14 performs post-processing on the image of the third region 43, which is the area where the metal cord layer 17 is exposed, in the same manner as the image of the second region 42 (step S304). The sizes of the second region 42 and the third region 43 quantitatively indicate the areas on the surface of the tire 30 that are extensively damaged.

[0080] Processor 14 performs post-processing on the image of the local damage portion 52 (such as a defect) output by inputting the tread image 40 into the second trained model (step S305). See reference... Figure 7The post-processing of the localized damage portion 52 may include calculating the center of gravity position of the localized damage portion 52 and calculating the area ratio of the area of ​​the localized damage portion 52 to the surface area of ​​the tread portion 35 of the tire 30. The processor 14 stores this information related to the center of gravity position and the area ratio in the storage unit 15.

[0081] Next, the processor 14 generates a mask 56 covering the second region 42, the third region 43, and the localized damage 52 on the tread image 40 (step S306). This mask 56 is similar to the mask 56 applied to the tread image 40 of the sample tire when generating the third training data. The mask 56 is used to exclude the second region 42, the third region 43, and the localized damage 52, which are considered to be areas of severe damage, from the evaluation target area in the following processing for determining relatively minor damage states such as scratches.

[0082] like Figure 10 In this way, processor 14 extracts a rectangular region 55 including the tread portion 35 of tire 30 from tread image 40 (step S307). By generating the segmented region in step S302, processor 14 can identify the extent of the tread portion 35. Processor 14 can also apply the mask 56 generated in step S306 to the rectangular region 55. In this way, the area other than the first region 41 exposed on the surface of the outer rubber 33 is masked.

[0083] Processor 14 evaluates the damage state of the surface area of ​​tire 30 (step S308). Specifically, processor 14 inputs the image of the rectangular region 55 extracted in step S307 into the third trained model and obtains numerical values ​​indicating the damage state of each sub-region 57 as output. Processor 14 can display the output numerical values ​​indicating the damage state of each sub-region 57 on the display included in the output unit 13 and store them in the storage unit 15.

[0084] As described above, the tire damage assessment device 10 can be based on Figure 13 The flowchart in the document is used to output the following assessment results, in which the damage status of the surface area of ​​the tread 35 is quantified. (1) Images used to indicate the exposure state of each layer in the multiple layers of the tread 35 of a tire 30 that is layered according to the wear of the tread 35, and quantitative data such as the area of ​​the exposed region of each layer. (2) An image for indicating local damage 52 (such as defects) occurring in the tread 35 of the tire 30, the number of local damage 52, the center of gravity of each local damage 52, and the area ratio of the local damage 52 in the tread 35. (3) An index used to quantify minor damage conditions (such as scratches) occurring in the surface area of ​​the outermost layer of rubber 33, which is the outermost layer of the tread 35. Thus, the tire damage assessment device 10 of this disclosure can quantify and assess the surface condition of the tread 35 of the tire 30. The tire damage assessment device 10 can display the assessment results on the display unit of the output unit 13 and / or transmit the assessment results to an external device via the communication unit 11. Therefore, the tire damage assessment device 10 can provide the user with the damage status of the surface area of ​​the tread 35 in a manner easily accessible to the user.

[0085] The processor 14 of the tire damage assessment device 10 can also estimate the wear characteristics of the tire 30 based on information already output about localized damage 52 (such as defects) and information about the damage state caused by scratches, etc. By using the tire damage assessment device 10, users can analyze the relationship between the material of the tire 30 and / or the shape of the tread pattern and tire wear. Furthermore, by using the information output by the tire damage assessment device 10, users can predict the lifespan of the tire 30.

[0086] (Other configuration examples of tire damage assessment systems) Figure 1 The illustrated tire damage assessment system 1 consists of a tire damage assessment device 10 and a camera device 20. Figure 14 Another configuration example of tire damage assessment system 1 is shown. Figure 14 In one embodiment, the tire damage assessment device 10 is located in a location far from the user (e.g., in the cloud). The tire damage assessment device 10 is configured to communicate with a portable information terminal 60 and an information processing device 70 on the user side via a network 80 such as the Internet.

[0087] Portable information terminal 60 includes, for example, smartphones and tablets. Information processing device 70 is a computer such as a PC. Similar to tire damage assessment device 10, portable information terminal 60 and information processing device 70 may include components such as communication units, input units, output units, processors, and storage units. Portable information terminal 60 may also include a camera unit.

[0088] The portable information terminal 60 can be configured to send a tread image 40 of the tire 30 captured by the camera unit to the tire damage assessment device 10. Therefore, the portable information terminal 60 is used as... Figure 1The tire damage assessment system 1 includes a camera device 20. Additionally, a portable information terminal 60 can acquire and execute applications corresponding to the tire damage assessment system 1. The user of the portable information terminal 60 can take images of the tire 30 according to the application's instructions and send those images to the tire damage assessment device 10.

[0089] The information processing device 70 can acquire the evaluation results from the tire damage assessment device 10 and display them on a display, which serves as an output unit. The information processing device 70 can store the evaluation results from the tire damage assessment device 10 in a storage unit. The information processing device 70 can use the evaluation results accumulated in the storage unit for further analysis. The information processing device 70 can use dedicated applications stored in the storage unit to analyze and evaluate the damage to the tire 30 while sending and receiving information with the tire damage assessment device 10.

[0090] Although embodiments according to this disclosure have been described with reference to the accompanying drawings and examples, it should be noted that various modifications or alterations can be readily made by those skilled in the art based on this disclosure. Therefore, it should be understood that such modifications or alterations are included within the scope of this disclosure. For example, the functions included in the constituent parts or steps may be rearranged in a manner that does not logically contradict each other, and multiple constituent parts or steps may be combined into one or divided. Embodiments according to this disclosure can also be implemented as methods, programs, or storage media recording such programs, executed by a processor equipped in a device. It should be understood that these are also included within the scope of this disclosure.

[0091] [Contribution to the UN-led Sustainable Development Goals (SDGs)] The SDGs have been promoted to achieve a sustainable society. One embodiment of the present invention may be a technology that contributes to, for example, "No. 9 Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation." Explanation of reference numerals in the attached figures

[0092] 10. Tire damage assessment equipment 11 Communication Unit 12 Input Units 13 Output Unit (Display Unit) 14 processors 15 storage units 20 camera devices 30 tires 31. Bead section 32 Side wall portion 33 Outer rubber layer 34 Inner rubber layer 35 Fetal face 36 fetuses 37. Metallic Cord Layer 38 Bead Core 40 Tread Image 41 First District 42 Second Region 43 Third Region 45 strip-shaped areas 51. Groove 52 Localized injury site 53 Repair Department 54 small areas 55 Rectangular area 56 Masks 57 sub-regions 60 Portable Information Terminal 70 Information processing equipment 80 Network

Claims

1. A tire damage assessment device, comprising a processor configured to acquire a tread image obtained by photographing the tread portion of a tire and to quantify the damage state of a surface region of the tread portion based on the tread image.

2. The tire damage assessment device according to claim 1, wherein, The tread portion includes multiple layers arranged radially along the tire, and the processor is configured to generate a segmented image in which a surface region of the tread portion included in the tread image is segmented using the layers exposed in that surface region.

3. The tire damage assessment device according to claim 2, wherein, The plurality of layers include an outer rubber layer constituting the tread surface, a metal cord layer disposed radially inward from the outer rubber layer along the tire, and an inner rubber layer located between the outer rubber layer and the metal cord layer.

4. The tire damage assessment device according to claim 2, wherein, The processor is configured to perform processing to: input a tire tread image into a first trained model, the first trained model being trained using first training data having multiple sample tire tread images as input and having a segmented image as output obtained by segmenting the surface regions of the sample tires included in the tread images using the layers exposed on the surface; and output the segmented image of the tire.

5. The tire damage assessment device according to claim 2, wherein, The processor is configured to divide a segmented image of the tire into a plurality of strip regions extending along the width direction of the tire, align the positions of the plurality of strip regions such that the tread portion of the strip regions is aligned in the width direction, and generate a histogram obtained by accumulating the pixels of the plurality of strip regions for each position in the width direction at least one layer exposed in the surface region.

6. The tire damage assessment device according to claim 1, wherein, The processor is configured to extract from the tread image at least one of the following: a groove portion indicating a groove in the tread area, a local damage portion indicating local damage to a surface area of ​​the tread area, and a repair portion indicating a location of previously occurring repair damage in the tire.

7. The tire damage assessment device according to claim 1, wherein, The processor is configured to perform processing to: input the tire tread image into a second trained model, the second trained model being trained using second training data having multiple sample tire tread images as input and having an image of a locally damaged portion indicating local damage to the tread portion of the sample tire included in the tread image as output; and output an image of the locally damaged portion of the tire.

8. The tire damage assessment device according to claim 7, wherein, The processor is configured to calculate the ratio of the area of ​​the localized damaged region to the area of ​​the surface region of the tummy rim.

9. The tire damage assessment device according to claim 7, wherein, The processor is configured to calculate the centroid position of the region of the localized damage.

10. The tire damage assessment device according to claim 2, wherein, The processor is configured to quantify the damage state of the surface region of the tire in the outermost exposed portion of the tire among the plurality of layers based on the tread image.

11. The tire damage assessment device according to claim 10, wherein, The processor is configured to extract an image of a surface region including the tread area from the tread image, divide the image into images of multiple sub-regions, and quantify the damage state of the surface region for each sub-region image.

12. The tire damage assessment device according to claim 10, wherein, The processor is configured to perform processing to: input the tread image of the tire into a third trained model, the third trained model being trained using third training data having tread images of multiple sample tires as input and having information as output for quantifying the damage state of the surface region in the outermost exposed portion of the tire among the multiple layers included in the tread image. And output information for quantifying the damage state of the surface region.

13. The tire damage assessment device according to claim 12, wherein, The damage state of the surface region in the third training data is quantified based on at least one of the number of scratches included in the surface region and the depth of the scratches.

14. The tire damage assessment device according to claim 1, wherein, The damage state includes the state of defects and scratches on the tire, and the processor is configured to estimate the wear characteristics of the tire based on information used to quantify the state of the defects and scratches.

15. The tire damage assessment device of claim 1, comprising at least one of a display unit and a communication unit, the display unit being configured to display information for quantifying the damage state of the surface area of ​​the tread portion, and the communication unit being configured to transmit the information to an external device.

16. A tire damage assessment method, performed by an information processing device, the tire damage assessment method comprising: Acquire images of the tire tread obtained by photographing the tire tread area; as well as The damage status of the surface area of ​​the tread region is quantified based on the tread image.

17. A tire damage assessment procedure configured to cause a computer to perform: processing to acquire a tread image obtained by photographing the tread portion of the tire; and processing to quantify the damage state of a surface region of the tread portion based on the tread image.

Images