Generating device for tire footprint image

Abstract

To obtain footprint images of a tire more easily than conventional arts.SOLUTION: A CPU 11 (processor) executes a processing process, a setting process, and an extraction process. In the processing process, the CPU 11 processes a planar tread surface image from an exterior image of a tire 100 including a curved tread surface 110. In the setting process, the CPU 11 obtains an extraction range of the planar tread surface image. In the extraction process, the CPU 11 extracts part of the planar tread surface images based on the extraction range. The extracted image is output as a footprint image. In addition, in the setting process, the CPU 11 determines a contact patch width and a contact patch length of the tire 100 based on specification data of the tire 100 and vehicle weight data. Further, in the setting process, the CPU 11 determines the extraction range based on the contact patch width and the contact patch length.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] This specification discloses an apparatus for generating a footprint image of a tire.

[0002] In Patent Document 1, analysis of a tire is performed based on the footprint of the tire. The footprint refers to the ground contact surface of the tire. Therefore, the footprint image is an image of the ground contact surface of the tire.

[0003] For example, grooves are formed on the tire surface. In the ground contact surface of the tire, a pipe structure is formed by the grooves of the tire and the road surface. In this pipe structure, noise is generated from the tire due to the occurrence of air column resonance. By acquiring the footprint image, the pipe structure such as the pipe length can be analyzed.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] By the way, when acquiring a footprint, using a measuring instrument makes the work process complicated. For example, an operator mounts a tire on a vehicle. Then, the operator operates the vehicle so that the tire rides on the measuring instrument. The measuring instrument is provided with a plurality of pressure sensors. The pressure distribution on the ground contact surface of the tire is obtained by the measuring instrument. This pressure distribution is used as a footprint image.

[0006] This specification discloses an apparatus for generating a footprint image of a tire, which can acquire the footprint image of the tire more simply than before.

Means for Solving the Problems

[0007] This specification discloses a device for generating a footprint image of a tire. This generating device includes a processor. The processor executes a processing process, a setting process, and an extraction process. In the processing process, the processor processes a planar tread surface image from an appearance image of a tire including a curved tread surface. In the setting process, the processor obtains an extraction range of the planar tread surface image. In the extraction process, the processor extracts a part of the planar tread surface image based on the extraction range. The extracted image is output as a footprint image. Also, in the setting process, based on the tire specification data and the vehicle weight data, the processor obtains the contact width and the contact length of the tire. Further, in the setting process, based on the contact width and the contact length, the processor determines the extraction range.

[0008] According to the above configuration, a footprint image is generated using the appearance image of the tire, the tire specification data, and the weight data of the vehicle on which the tire is mounted as input data.

[0009] Also, in the above configuration, in the processing process, the processor performs the following processing. That is, the processor obtains the center line of the appearance image of the tire. Further, the processor divides the appearance image of the tire using the center line as a dividing line. Next, the processor performs a projective transformation on each of the divided images. Further, the processor generates a planar tread surface image by combining the divided images after the projective transformation.

[0010] According to the above configuration, the curved tread surface is converted into a planar shape. That is, an image similar to the tread surface on the contact surface can be generated by the projective transformation.

[0011] Also, in the above configuration, in the setting process, the processor performs the following processing. That is, the processor generates an extraction frame based on the contact width and the contact length. Further, the processor performs an R process on the extraction frame.

[0012] According to the above configuration, an image closer to the actual footprint image can be generated.

[0013] Also, in the above configuration, in the projective transformation, the processor sets the vertices of the projective transformation. In this setting, the processor sets corresponding points on the tangent line of the boundary edge of the curved tread surface included in the divided image before the transformation.

[0014] According to the above configuration, a projective transformation corresponding to the curved surface of the tread surface becomes possible.

[0015] Also, in the above configuration, a ground contact length neural network and a ground contact width neural network are implemented in the processor. The ground contact length neural network takes tire specification data and vehicle weight data as the input layer and the ground contact length as the output layer. The ground contact width neural network takes tire specification data and vehicle weight data as the input layer and the ground contact width as the output layer.

[0016] According to the above configuration, the ground contact length neural network and the ground contact width neural network can be trained using data related to the footprint image accumulated conventionally as teacher data.

Effect of the Invention

[0017] According to the tire footprint image generation device disclosed in this specification, a tire footprint image can be obtained more simply than in the prior art.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Embodiments for Carrying Out the Invention

[0019] 1. Hardware Configuration The tire footprint image generation device 10 according to the present embodiment is exemplified in FIG. 1. Hereinafter, the tire footprint image generation device 10 is appropriately described as the "image generation device 10".

[0020] The image generation device 10 is composed of, for example, a computer. The image generation device 10 includes a CPU 11, a RAM 12, a ROM 13, a storage 14, an input / output controller 15, and a display unit 16.

[0021] The CPU 11 is a central processing unit, also called a processor. The RAM 12 is a volatile storage device that temporarily stores data during operation. The ROM 13 is a storage device from which data can be read. The storage 14 is a storage device capable of writing and reading data. The storage 14 is composed of, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0022] 2. Functional Blocks When the CPU 11 (processor) executes a program stored in the ROM 13 or the storage 14, functional units as illustrated in FIG. 2 are configured in the image generation device 10. That is, the image generation device 10 includes an image processing unit 20, a ground length CNN 21, a ground width CNN 22, and an image extraction unit 23 as functional units. In other words, these functional units are implemented in the CPU 11. That is, the CPU 11 can execute the processing by the above functional units. The processing content by the above functional units will be described later.

[0023] Data is input to the image generation device 10 from the input device 17 and the imager 19. The input device 17 is, for example, a keyboard. Also, as will be described later, external appearance image data of a tire is input to the image generation device 10 from the imager 19.

[0024] 3. Process for Generating a Footprint Image FIG. 3 illustrates a generation flow of a footprint image. This generation flow is executed by the image generation device 10. This generation flow is branched into two processes. That is, the generation flow is branched into a processing process (S10 - S17) and a setting process (S20 - S25). The processing process and the setting process merge at a later stage. Further, the generation flow executes an extraction process (S30, S31).

[0025] In the processing operation, the curved tread surface image is processed into a planar shape. In the setting operation, the extraction range of the planar tread surface image is determined. Further, in the extraction operation, a part of the planar tread surface image is extracted based on the extraction range. The extracted image is output as a footprint image.

[0026] 3-1. Flattening of the Tread Surface Image (Processing Operation) Referring to FIG. 2, an external appearance image of the tire is input to the image generation device 10 from the imager 19. FIG. 4 illustrates an external appearance image of the tire 100 captured by the imager 19. This external appearance image includes the curved tread surface 110. That is, the non-grounded tread surface 110 is included in the captured image.

[0027] The external appearance of the tire 100 is imaged in a state where no pressure is applied to the tread surface 110. For example, as shown in FIG. 4, the tire 100 is imaged in a laid-down state. As shown in FIG. 4, the external appearance image of the tire 100 has an arc-shaped contour from the longitudinal center toward both ends. By the image generation device 10, an image of such a curved tread surface 110 is processed into a planar image (see FIG. 10).

[0028] Referring to FIGS. 2 and 3, the image processing unit 20 requests an external appearance image of the tire 100 (S10). For example, the image processing unit 20 displays an image request message on the display unit 16. Further, the image processing unit 20 checks whether an external appearance image of the tire 100 has been input (S11). If no image has been input, the flow returns to step S10.

[0029] When an external appearance image of the tire 100 is input, the image processing unit 20 deletes the background from the image (S12). For example, the background image is deleted using the Grab Cut Algorithm.

[0030] The GrabCut algorithm is a well-known image processing technique. Therefore, the GrabCut algorithm will be briefly described below. First, a foreground region is specified from an appearance image of the tire 100 (see FIG. 4). For example, an operator of the input device 17 specifies the foreground region. In the present embodiment, the foreground region is an image region of the tire 100. Then, based on the specification of the foreground region, the image processing unit 20 divides the foreground region and the background region.

[0031] The foreground image after the division process is displayed on the display unit 16. The operator of the input device 17 checks the displayed image. When the foreground is included in the background or the background is included in the foreground, the input device 17 designates the background region during foregrounding and the foreground region in the background. By repeating such interactive extraction processing, the background is deleted from the image of FIG. 4. The tire image in the deletion process of step S12 is illustrated in FIG. 5.

[0032] Next, the image processing unit 20 binarizes the tire image of FIG. 5 (S13). In the binarization process, the pixel value of a pixel on the tire image is divided into either 0 (black) or 255 (white). FIG. 6 illustrates the binarized tire image. For example, the tire image is binarized by adaptive binarization. Adaptive binarization is a well-known binarization technique. Therefore, the process by adaptive binarization will be briefly described below.

[0033] In adaptive binarization, the average value of a conversion pixel and its surrounding pixels becomes the binarization threshold. For example, assume that a certain pixel (pixel A1) and another pixel (pixel A2) have the same pixel value (for example, 120). When the average value of the pixel values of pixel A1 and its surrounding pixels is lower than pixel A1 (for example, 100), pixel A1 is converted to white (120 → 255). Also, when the average value of the pixel values of pixel A2 and its surrounding pixels is higher than pixel A2 (for example, 140), pixel A2 is converted to black (120 → 0). In this way, the binarization threshold is set adaptively according to the brightness of the surroundings, etc. By binarizing the tire image, geometric operations as described below can be performed with high accuracy.

[0034] The image processing unit 20 obtains the center line of the binarized tire image (S14). In FIGS. 7, 9, 10, and 13-16, the X-Y coordinate system is used as the coordinate system of the image. The X-axis and the Y-axis are orthogonal.

[0035] Referring to FIG. 7, for example, the image processing unit 20 obtains the end point P1 of the binarized image of the tire 100 on the longitudinal axis (X-axis). For example, the image processing unit 20 extracts a pixel with the maximum X-axis coordinate value from the pixels with a pixel value of 255 (white), and sets the pixel as the end point P1. Similarly, the image processing unit 20 extracts a pixel with the minimum X-axis coordinate value from the pixels with a pixel value of 255 (white), and sets the pixel as the end point P2.

[0036] Furthermore, the image processing unit 20 obtains the center point P3 of the straight line L1 connecting the end points P1 and P2. Next, the image processing unit 20 obtains a perpendicular line from the center point P3 to the straight line L1. This perpendicular line becomes the center line L2.

[0037] Referring to FIG. 8, next, the image processing unit 20 divides the appearance image of the tire into images F1 and F2 using the center line L2 as a dividing line (boundary line) (S15). Next, the image processing unit 20 performs projective transformation on each of the divided images F1 and F2 (S16).

[0038] FIG. 9 shows an example of projective transformation for the divided image F1. Projective transformation is also called homography transformation. In projective transformation, conversion from a quadrilateral to another quadrilateral is performed. For example, conversion from a trapezoid to a rectangle is performed.

[0039] In homography transformation, the four vertex coordinates (x11, y11), (x12, y12), (x13, y13), (x14, y14) of the quadrilateral move to (x21, y21), (x22, y22), (x23, y23), (x24, y24). The movement of this vertex is expressed as the following mathematical formula (eq. 1).

[0040]

Equation

[0041] The vertices (x11, y11), (x12, y12), (x13, y13), (x14, y14) before transformation and the vertices (x21, y21), (x22, y22), (x23, y23), (x24, y24) after transformation are specified in advance by an operator or the like. Therefore, the mathematical formula (eq. 1) becomes an eight-variable equation including the homography parameters h11, h12, h13, h21, h22, h23, h31, h32. Also, in the mathematical formula (eq. 1), eight mathematical formulas are represented in matrix form. That is, by substituting the numerical values of (x11, y11), (x12, y12), (x13, y13), (x14, y14), (x21, y21), (x22, y22), (x23, y23), (x24, y24) into the mathematical formula (eq. 1), the homography parameters h11, h12, h13, h21, h22, h23, h31, h32 can be calculated.

[0042] FIG. 9 illustrates a process of determining four vertices in projective transformation. Vertices C and D are set on the center line L2 which is the center of the tire. The image processing unit 20 searches for pixels with a pixel value of 255 (white) among the pixels on the center line L2. Further, among these pixels, the image processing unit 20 sets the uppermost point and the lowermost point on the center line L2 as vertices C and D. Vertices C and D are fixed points that do not move before and after projective transformation. That is, for vertex C, the relationship (x11, y11) = (x21, y21) holds. Also, for vertex D, the relationship (x12, y12) = (x22, y22) holds.

[0043] Next, the image processing unit 20 approximately represents the boundary edge of the curved tread surface 110 as a curve in the divided image F1. In other words, the image processing unit 20 approximately represents the boundary line between the tread surface 110 and the background as a curve. Since the tread surface 110 has boundaries with the background above and below, two approximate curves can be obtained.

[0044] The two approximate curves intersect the vertex C and the vertex D. The image processing unit 20 obtains the tangent line L10 of the approximate curve at the vertex C. Similarly, the image processing unit 20 obtains the tangent line L12 of the approximate curve at the vertex D. The tangent lines L10 and L12 can be said to be the tangent lines of the boundary edge of the tread surface 110. The vertices A and B are set on the tangent lines L10 and L12. For example, the intersections of the tangent lines L10 and L12 with the side edges of the divided image F1 are the vertices A and B.

[0045] Next, the image processing unit 20 obtains perpendicular lines L13 and L14 to the center line L2. Here, the perpendicular lines L13 and L14 intersect the vertex C and the vertex D. The intersections of the perpendicular lines L13 and L14 with the side edges of the divided image F1 are the vertices A' and B'. Further, the image processing unit 20 obtains the homography parameters for moving the vertices A and B to the vertices A' and B' with the vertices C and D as fixed points.

[0046] In this way, by performing projective transformation according to the curved surface shape of the tread surface 110, the planarization of the tread surface 110 can be accurately performed.

[0047] The projective transformation illustrated in FIG. 9 is also executed for the divided image F2 (see FIG. 8). The image processing unit 20 combines the divided images F1' and F2' after the projective transformation. FIG. 10 shows the planar tread surface image F10, which is the combined image.

[0048] 3-2. Determination (Setting Process) of Extraction Frame Referring to FIG. 2, a ground length CNN21 (ground length neural network) and a ground width CNN22 (ground width neural network) are implemented in the CPU 11 of the image generation device 10. CNN is an abbreviation for Convolutional Neural Network.

[0049] FIG. 11 illustrates the ground length CNN21. The ground length CNN21 includes an input layer, a hidden layer, and an output layer. A plurality of nodes are provided in each layer.

[0050] In the input layer, the specification data of the tire 100 and the vehicle weight data are input. The specification data includes data such as the dimensions, weight, material, and functions of the tire 100. For example, the specification data of the tire 100 includes tire weight, tread rubber hardness, tire size, tire weight, rolling resistance coefficient, tread rubber material, load index, static load radius, aspect ratio, rim diameter, and rim width. The specification data of the tire 100 can be obtained from the supplier of the tire 100.

[0051] Also, the vehicle weight data can be obtained from the vehicle supplier. Regarding the vehicle weight data, instead of inputting the numerical value described in the specification table, a numerical value obtained by reducing the said numerical value by a predetermined ratio may be input.

[0052] From the output layer, the contact length of the tire 100 is output. A plurality of nodes are provided in the output layer. From each node, the contact length and the accuracy are output.

[0053] In FIG. 12, the contact width CNN22 is illustrated. Similar to the contact length CNN21, the contact width CNN22 includes an input layer, a hidden layer, and an output layer. A plurality of nodes are provided in each layer.

[0054] Similar to the contact length CNN21, in the input layer of the contact width CNN22, the specification data of the tire 100 and the vehicle weight data are input. From the output layer, the contact width of the tire 100 is output. A plurality of nodes are provided in the output layer. From each node, the contact width and the accuracy are output.

[0055] In the conventional tire characteristic analysis, the specification data of the tire and the vehicle weight, and the contact length and the contact width are recorded. Therefore, teacher data can be created from the past tire characteristic analysis results. In this teacher data, the input data is the specification data of the tire and the vehicle weight. Also, the correct answer data is the contact length and the contact width. The learned contact length CNN21 and the contact width CNN22 are implemented in the CPU 11 (see FIG. 1) of the image generation device 10 by such teacher data.

[0056] Referring to FIG. 3, the image generation device 10 requests the specification data of the tire 100 (S20). For example, the image generation device 10 displays a specification data request message on the display unit 16 (see FIG. 2). The specification data of the tire 100 is input from the input device 17. Further, the image generation device 10 checks whether the specification information of the tire 100 has been input (S21). If the specification information has not been input, the flow returns to step S20.

[0057] When the specification information of the tire 100 is input, the image generation device 10 requests the vehicle weight data (S22). For example, the image generation device 10 displays a vehicle weight data request message on the display unit 16. The vehicle weight data is input from the input device 17. Further, the image generation device 10 checks whether the vehicle weight information has been input (S23). If the weight information has not been input, the flow returns to step S22.

[0058] When the specification data of the tire 100 and the vehicle weight data are input, these data are input to the input layers of the contact length CNN 21 and the contact width CNN 22 (S24). The image extraction unit 23 acquires the values of the contact length and the contact width with the highest accuracy from the output layers of the contact length CNN 21 and the contact width CNN 22 (S25). Based on the acquired contact length and contact width, the extraction range of the image is determined.

[0059] 3-3. Extraction of Footprint Image (Extraction Process) Referring to FIGS. 2 and 13, the image processing unit 20 obtains the distance between vertex C and vertex D from the tire size. For example, the tire size is expressed as "250 / 45 / R18 100W". According to this example, the tire width is 250 mm. The image processing unit 20 sets the distance between vertex C and vertex D to the tire width (for example, CD = 250 mm).

[0060] The image processing unit 20 determines the size of the extraction frame FL1 based on the value of the tire width. The horizontal dimension (X-axis length) and vertical dimension (Y-axis length) of the extraction frame FL1 are the contact length and contact width obtained in step S25. For example, the image processing unit 20 obtains the lengths (scales) of the contact length L31 and contact width L30 on the image when the distance between the vertex C and the vertex D is taken as the tire width.

[0061] Referring further to FIG. 13, the image processing unit 20 obtains the center point P10. The center point P10 is the midpoint of the vertex C and the vertex D. The image processing unit 20 aligns the center of the extraction frame FL1 with the center point P10.

[0062] Based on the extraction frame FL1, the image processing unit 20 extracts a part of the planar tread surface image F10 (S30). That is, as illustrated in FIG. 14, among the planar tread surface images F10, the image area within the extraction frame FL1 is extracted.

[0063] Here, in FIG. 14, the extracted image is rectangular. On the other hand, the actual footprint image has rounded corners. Therefore, as illustrated in FIG. 15, R processing may be performed on the extraction frame FL1. For example, among the contact width L30 and contact length L31, the value of 1 / 4 of the shorter one is set as the radius of curvature.

[0064] The image extracted by the extraction frame FL1 subjected to R processing is illustrated in FIG. 16. The image processing unit 20 outputs the extracted image as the footprint image 60 (S31). For example, the footprint image 60 is displayed on the display unit 16 (see FIG. 2).

[0065] In the above-described embodiment, the captured image by the imager 19 is used as the appearance image of the tire 100. Instead of the captured image, 3D data by 3D-CAD or the like may be input as the appearance image of the tire 100.

Explanation of Reference Numerals

[0066] 10 Image generation device, 11 CPU (processor), 20 Image processing unit, 21 Ground length CNN, 22 Ground width CNN, 23 Image extraction unit, 60 Footprint image, 100 Tire, 110 Tread surface.

Claims

1. An apparatus for generating a footprint image of a tire, comprising a processor, wherein the processor performs a processing operation to process a planar tread surface image from an external appearance image of a tire including a curved tread surface, a setting operation to determine an extraction range of the planar tread surface image, an extraction operation to extract a part of the planar tread surface image based on the extraction range and output it as a footprint image, and executes the processor in the setting operation, obtains the contact width and contact length of the tire based on the tire specifications data and vehicle weight data, and further in the setting operation, determines the extraction range based on the contact width and the contact length, An apparatus for generating a footprint image of a tire.

2. The apparatus for generating a footprint image of a tire according to claim 1, wherein in the processing operation, the processor obtains a center line of the external appearance image of the tire, divides the external appearance image of the tire using the center line as a dividing line, performs a projective transformation on each of the divided images, and generates the planar tread surface image by combining the divided images after the projective transformation, An apparatus for generating a footprint image of a tire.

3. The apparatus for generating a footprint image of a tire according to claim 2, wherein in the setting operation, the processor generates an extraction frame based on the contact width and the contact length, and performs R processing on the extraction frame, An apparatus for generating a footprint image of a tire.

4. The apparatus for generating a footprint image of a tire according to claim 2 or 3, wherein in the projective transformation, the processor sets the vertex of the projective transformation on the tangent line of the boundary edge of the curved tread surface included in the divided image before transformation, An apparatus for generating a footprint image of a tire.

5. The apparatus for generating a footprint image of a tire according to claim 1, wherein the processor is implemented with a contact length neural network having the tire specifications data and vehicle weight data as an input layer and the contact length as an output layer, a contact width neural network having the tire specifications data and vehicle weight data as an input layer and the contact width as an output layer, An apparatus for generating a footprint image of a tire. ​

Images