JP2024037490A - Tire grounding shape analysis device and tire grounding shape analysis method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To acquire a proper grounding shape when acquiring a digital image by a camera.

SOLUTION: A tire grounding shape analysis device comprises: an image acquisition unit which acquires a grounding surface image of a tire; a temporary GCA calculation unit which obtains a temporary GCA image by performing binarization processing with a prescribed threshold as a reference for the acquired grounding surface image; a vertical line installation processing unit which installs a vertical line in each pixel on a boundary line for the temporary GCA image; a luminance differentiation processing unit which differentiates the distribution of luminance along the vertical line for each installed vertical line and obtains a maximum point and a minimum point by performing further differentiation; a boundary setting processing unit which sets a boundary in the temporary GCA image on the basis of the obtained maximal point and minimum point; a smoothing processing unit which smooths the boundary for the temporary GCA image set with the boundary; and a superimposition processing unit which obtains the grounding shape by superimposing a groove image obtained from the grounding surface image on the temporary GCA image with the smoothed boundary.

SELECTED DRAWING: Figure 2

COPYRIGHT: (C)2024,JPO&INPIT

Description

The present invention relates to a tire ground contact shape analysis device and a tire ground contact shape analysis method.

2. Description of the Related Art A technique is known in which a digital image of a tire footprint is obtained by photographing the contact surface of a tire on the upper side of a transparent plate with a camera from below the transparent plate (for example, Patent Document 1). Furthermore, a technique is known in which the contact surface of a tire is irradiated with light and photographed with a camera from below a transparent plate (for example, Patent Document 2).

Patent No. 3293670 JP2020-019437A

However, if a certain threshold value is used when acquiring a digital image by photographing with a camera, it may not be possible to acquire an appropriate ground shape image.

The present invention has been made in view of the above, and an object of the present invention is to obtain a tire ground contact shape analysis device and a tire ground contact shape analysis method that can obtain an appropriate ground contact shape when capturing a digital image by photographing with a camera. The goal is to provide the following.

In order to solve the above-mentioned problems and achieve the objects, a tire ground contact shape analysis device according to an aspect of the present invention includes an image acquisition unit that acquires a tire contact patch image, and a contact patch image acquired by the image acquisition unit. , a provisional GCA calculation unit that performs binarization processing based on a predetermined threshold value to obtain a provisional GCA image that is a provisional total ground contact area image; Luminance differentiation processing for differentiating the luminance distribution along the vertical line for each of the vertical lines installed by the vertical line installation processing unit and the vertical line installation processing unit, and further differentiating to obtain local maximum points and minimum points. a boundary setting processing section that sets a boundary in the temporary GCA image based on the maximum point and minimum point obtained by the luminance differentiation processing section; and a boundary setting processing section for setting the boundary in the temporary GCA image by the boundary setting processing section. Regarding the image, a smoothing processing unit smoothes the boundary, and a groove image obtained from the ground contact surface image is superimposed on the provisional GCA image whose boundary has been smoothed by the smoothing processing unit to obtain a ground contact shape. An overlay processing section.

A tire contact shape analysis method according to an aspect of the present invention includes an image acquisition step of acquiring a tire contact surface image, and a binarization process based on a predetermined threshold value on the contact surface image acquired in the image acquisition step. a provisional GCA calculation step for obtaining a temporary GCA image that is a provisional total ground area image; a vertical line installation processing step for installing a vertical line at each pixel on a boundary line for the temporary GCA image; and a vertical line installation processing step for installing a vertical line at each pixel on the boundary line. A brightness differentiation processing step for differentiating the brightness distribution along each vertical line set in the step and further differentiating it to obtain a local maximum point and a local minimum point; and a local maximum point obtained by the brightness differentiation processing step. and a boundary setting process step of setting a boundary in the temporary GCA image based on the minimum point; and a smoothing process step of smoothing the boundary of the temporary GCA image whose boundary has been set in the boundary setting process step. , a superposition processing step of superimposing a groove image obtained from the ground contact surface image on the provisional GCA image whose boundary has been smoothed in the smoothing processing step to obtain a ground contact shape.

According to the present invention, when acquiring a digital image by photographing with a camera, an appropriate ground contact shape can be acquired.

FIG. 1 is a configuration diagram schematically showing a tire ground contact shape analysis device according to an embodiment. FIG. 2 is a block diagram showing the functions of the tire ground contact shape analysis device shown in FIG. FIG. 3 is a diagram illustrating the movement of the transparent plate and the operation of the trigger device. FIG. 4 is a diagram illustrating the movement of the transparent plate and the operation of the trigger device. FIG. 5 is a flowchart showing the operation of the tire ground contact shape analysis device. FIG. 6 is a diagram illustrating the process of extracting grooves using vertical lines. FIG. 7 is a flow diagram showing the process of extracting grooves using vertical lines. FIG. 8 is a flow diagram showing the procedure of groove extraction processing. FIG. 9 is a diagram in which vertical lines are added to the grooves on the ground plane image. FIG. 10 is a diagram showing an example of two maximum brightness positions. FIG. 11 is a diagram illustrating the content of processing by the provisional GCA calculation unit. FIG. 12 is a diagram illustrating the arrangement of each illumination lamp. FIG. 13 is a diagram illustrating the arrangement of each illumination lamp. FIG. 14 is a diagram illustrating the arrangement of each illumination lamp. FIG. 15 is a diagram illustrating the arrangement of each illumination lamp. FIG. 16 is a diagram showing an example of a median filter. FIG. 17 is an explanatory diagram of the contraction process. FIG. 18 is an explanatory diagram of the expansion process. FIG. 19 is a flowchart showing the processing contents of the vertical line installation processing section. FIG. 20 is a diagram illustrating a procedure for installing a vertical line through the processing of the vertical line installation processing unit. FIG. 21 is a diagram illustrating the processing of the luminance differential processing section. FIG. 22 is a diagram illustrating the process of determining brightness by linear interpolation. FIG. 23 is a diagram illustrating the process of smoothing the ground plane image. FIG. 24 is a diagram showing an example of brightness distribution. FIG. 25 is a diagram showing an example of a ground plane image. FIG. 26 is a diagram showing the luminance distribution along the vertical line shown in FIG. 25. FIG. 27 is an erroneous determination map when boundary determination processing is performed on FIG. 26. FIG. 28 is an erroneous determination map when the position of the first local maximum point in FIG. 26 is determined to be a boundary. FIG. 29 is a diagram showing the luminance distribution along the vertical line shown in FIG. 25. FIG. 30 is an erroneous determination map when the position of the first minimum point in FIG. 29 is determined to be a boundary. FIG. 31 is an erroneous determination map when the position of the second local maximum point in FIG. 29 is determined to be a boundary. FIG. 32 is an erroneous determination map when the boundary in FIG. 29 is determined based on other conditions. FIG. 33 is a diagram showing the luminance distribution along the vertical line shown in FIG. 25. FIG. 34 is an erroneous determination map when boundaries are determined based on other conditions. FIG. 35 is a diagram illustrating the procedure for determining the contour. FIG. 36 is a diagram showing a state in which vertical lines are set for each pixel. FIG. 37 is a diagram showing an example of the coordinates of the boundary position. FIG. 38 is a diagram illustrating an example of a boundary determination result before contour smoothing. FIG. 39 is a diagram showing an example of a boundary determination result after contour smoothing. FIG. 40 is a diagram illustrating the overlapping process. FIG. 41 is a diagram illustrating the length of a vertical line set by the vertical line installation processing section. FIG. 42 is a diagram illustrating an example of the result of binarizing the ground plane image in the temporary GCA calculation unit. FIG. 43 is a diagram illustrating an example of the result of binarizing the ground plane image in the temporary GCA calculation unit.

Embodiments of the present invention will be described in detail below based on the drawings. In the following description of each embodiment, the same or equivalent components as in other embodiments will be denoted by the same reference numerals, and the description thereof will be simplified or omitted. The present invention is not limited to each embodiment. Furthermore, the constituent elements of each embodiment include those that can be easily replaced by those skilled in the art, or those that are substantially the same. Note that the configurations described below can be combined as appropriate. Further, the configuration may be omitted, replaced, or changed without departing from the gist of the invention.

FIG. 1 is a configuration diagram schematically showing a tire ground contact shape analysis device 1 according to an embodiment. FIG. 2 is a block diagram showing the functions of the tire ground contact shape analysis device 1 shown in FIG. In these figures, FIG. 1 schematically shows the overall configuration of the tire ground contact shape analysis device 1, and FIG. 2 shows the main functions of the tire ground contact shape analysis device 1.

The tire ground contact shape analysis device 1 according to the present embodiment is applied to a system that analyzes the ground contact surface 61 of a pneumatic tire 60 by acquiring an image of the ground contact surface 61. The tire contact profile analysis device 1 includes a tire testing machine 2, an imaging device 10, and a tire contact patch analysis device 20.

The tire testing machine 2 is a device that applies test conditions to a pneumatic tire 60 (hereinafter referred to as tire 60) that is an analysis target. In the configuration of FIG. 1, the tire testing machine 2 includes a support device 3, a drive device 5, and a transparent plate 11. The support device 3 is a device that rotatably supports the tire 60, and has a rim 4 on which the tire 60 is mounted. The drive device 5 is a device that applies driving force to the tires 60 and the transparent plate 11. The drive device 5 includes a motor 6 that drives the tires 60 and the transparent plate 11, and a motor control device 7 that controls the motor 6. Note that the drive device 5 includes a gear (not shown) and drives the transparent plate 11 horizontally.

In this tire testing machine 2, the support device 3 supports the tire 60 mounted on the rim 4, and the tire 60 is pressed against the upper surface 11U, which is one main surface of the transparent plate 11, to apply a load to the tire 60. The transparent plate 11 reproduces a flat road surface. The tire 60 pressed against the transparent plate 11 has its ground contact surface 61 deformed in the same way as when the tire 60 is running on a flat road surface. By horizontally driving the transparent plate 11, the rolling state of the tires 60 when the vehicle is running is reproduced using the surface of the transparent plate 11 as a road surface, and dynamic ground contact characteristics can be analyzed. The drive device 5 can rotate the rim 4 by a predetermined angle by driving the motor 6 using the motor control device 7.

The transparent plate 11 is a light transmitting plate that has the property of transmitting light. The transparent plate 11 does not need to transmit 100% of the light, but only needs to have a light transmittance that allows the surface of the tire 60 to be photographed through the transparent plate 11. The transparent plate 11 is, for example, a flat plate made of acrylic resin or a flat plate made of glass. Since the contact state between the tire 60 and the flat plate is photographed and image analyzed, a more realistic ground contact state of the tire 60 can be analyzed. Regarding the transparent plate 11, there are no specifications such as plate thickness and refraction angle.

The photographing device 10 includes a camera 15 that is a photographing unit that photographs the tire 60, an illumination lamp 16 that is a light source, and a trigger device 17. The camera 15 is configured by, for example, a CCD (Charge Coupled Device) camera. The camera 15 is fixed within the photographing device 10. The camera 15 photographs the tire 60 through the transparent plate 11, thereby photographing the ground contact surface 61 of the tire 60 pressed against the transparent plate 11. Specifically, the camera 15 is disposed on the lower surface 11D side, which is the other main surface of the transparent plate 11, with its optical axis perpendicular to the lower surface 11D side, and the camera 15 is installed on the lower surface 11D side, which is the other main surface of the transparent plate 11, and the camera 15 is arranged so that the optical axis is perpendicular to the lower surface 11D side. Shoot 60. Thereby, the camera 15 photographs the tire 60 including at least the contact patch 61, and generates digital image data of the tire 60 including the contact patch 61.

The illumination lamp 16 is a lamp that illuminates the photographing range of the camera 15, and is composed of, for example, a halogen lamp. The illumination lamp 16 is a general term for a plurality of lamps 161 to 168, which are provided as described later. The illumination lamp 16 irradiates the contact surface 61 of the tire 60 pressed against the transparent plate 11 with light. The illumination lamp 16 emits light from the lower surface 11D side of the transparent plate 11 through the transparent plate 11 or from between the upper surface 11U side of the transparent plate 11 and the tire 60. The plurality of illumination lamps 16 are arranged at positions other than the position where the transparent plate 11 moves. Note that as the photographing device 10 moves, the camera 15 and illumination lamp 16 in the photographing device 10 move together.

Note that the number of these illumination lamps 16 may be varied depending on the test conditions in the tire testing machine 2. For example, when the load when pressing the tire 60 against the transparent plate 11 is small, the contact area becomes narrower, and the difference in brightness between the contact surface 61 and the non-contact surface becomes clear. Therefore, in this case, the number of illumination lamps 16 may be relatively small, and may be arranged at two locations diagonally to the moving direction of the transparent plate 11. On the other hand, if the load when pressing the tire 60 against the transparent plate 11 is large, the ground contact area becomes wider, so it is necessary to irradiate the ground contact surface 61 with light from more directions. Therefore, in this case, the illumination lamps 16 are arranged at four or more locations surrounding the ground plane 61. Further, these illumination lamps 16 may be of a constant lighting type or a flash lighting type.

The trigger device 17 is a device that outputs a trigger signal indicating the timing of photographing by the camera 15. The trigger device 17 outputs a semiconductor laser and outputs a trigger signal when the reflected light is detected. In this example, a retroreflection sheet 18 is attached to the side surface of the transparent plate 11, and the semiconductor laser outputted by the trigger device 17 is reflected by the retroreflection sheet 18, and the detection unit 171 of the trigger device 17 detects the reflected light. Outputs a trigger signal when detected. If the relationship between the pasting position of the retroreflective sheet 18 and the position of the camera 15 is fixed, it is possible to photograph the ground contact surface 61 of the tire 60 at the same position even when photographing is performed a plurality of times.

The tire contact surface analysis device 20 is, for example, a PC (Personal Computer) installed with a predetermined analysis program, and processes an image of the tire 60 input from the imaging device 10 to analyze the contact surface 61 of the tire 60. I do. The process of analyzing the ground contact surface 61 of the tire 60 includes the process of calculating the ground contact surface 61 based on the photographed image of the tire 60. The tire contact patch analysis device 20 includes a processing device 30 that performs arithmetic processing such as analysis of the contact patch 61 and data storage, an input unit 21 that allows an operator to perform input operations to the tire contact patch analysis device 20, and an input unit 21 that performs an input operation to the tire contact patch analysis device 20. and a display section 22 that displays various information. The input section 21 uses a keyboard, a pointing device such as a mouse, and the display section 22 uses a display device such as a liquid crystal display. The input section 21 and the display section 22 are electrically connected to the processing device 30, so that the tire contact surface analysis device 20 allows the operator to perform input operations on the input section 21 while visually checking the display section 22. is now possible. Further, the camera 15 is connected to the processing device 30 of the tire contact surface analysis device 20, so that the tire contact surface analysis device 20 can acquire images taken by the camera 15.

The processing device 30 included in the tire contact surface analysis device 20 includes a processing section 31 having a CPU (Central Processing Unit) and the like, and a storage section 32 such as a RAM (Random Access Memory). The processing unit 31 and the storage unit 32 configured in this manner may be provided in the same housing, or may be provided in different housings, or a plurality of storage units 32 may be provided in both forms. may be provided.

The processing unit 31 included in the processing device 30 includes an image acquisition unit 33 , a temporary GCA calculation unit 34 , a vertical line installation processing unit 35 , a luminance differentiation processing unit 36 , a boundary setting processing unit 37 , and a smoothing processing unit 38 , an overlay processing section 39 , and a groove image extraction processing section 40 . Furthermore, the processing unit 31 creates a contact area image indicating the contact area of the tire 60 based on the contact area image acquired by the image acquisition unit 33. In the following description, all images are assumed to have 256 gradations, and black is defined as a brightness of "255" and white as a brightness of "0".

Here, the grooves of the tire 60 are a general term for the main grooves, sub-grooves, and chamfers provided at the openings of these grooves provided on the tread surface of the tire 60. The main groove is a groove that is required to display a wear indicator as defined by JATMA. Moreover, the sub-groove is a lateral groove extending in the width direction of the tire, and functions as a groove by opening when the tire contacts the ground.

The image acquisition unit 33 acquires a tire contact surface image. The ground plane image is a digital image taken by the camera 15.

The provisional GCA calculation unit 34 calculates a provisional GCA image by performing a binarization process on the ground contact surface image acquired by the image acquisition unit 33 using a predetermined threshold as a reference. The provisional GCA image is not an accurate ground contact area image (hereinafter sometimes abbreviated as GCA) but a provisional total ground contact area image.

The vertical line installation processing unit 35 performs a process of installing a vertical line at each pixel on the boundary line in the temporary GCA image calculated by the temporary GCA calculation unit 34. The brightness differentiation processing section 36 differentiates the distribution of brightness along each vertical line set by the vertical line installation processing section 35, and further differentiates the distribution to find a maximum point and a minimum point.

The boundary setting processing unit 37 sets boundaries in the provisional GCA image based on the maximum points and minimum points obtained by the luminance differentiation processing unit 36. The smoothing processing unit 38 performs a process of smoothing the boundaries of the temporary GCA image whose boundaries have been set by the boundary setting processing unit 37.

The superposition processing unit 39 superimposes the groove image obtained from the ground contact surface image on the provisional GCA image whose boundaries have been smoothed by the smoothing processing unit 38 to obtain the ground contact shape. The groove image extraction processing section 40 performs a process of obtaining a groove image to be used in the superimposition processing section 39 based on the ground contact surface image.

The storage unit 32 stores in advance an analysis program used by the tire contact surface analysis device 20. When acquiring the ground contact characteristics of the contact surface 61 of the tire 60, the processing unit 31 calls a program stored in the storage unit 32, and executes operations according to the program to perform each function. Execute.

The tire ground contact shape analysis device 1 according to this embodiment has the above configuration. Hereinafter, the operation of the tire ground contact shape analysis device 1 will be explained. When analyzing the contact surface 61 of the tire 60 using the tire contact profile analysis device 1, the tire 60 is mounted on the support device 3 of the tire testing machine 2, and the tire 60 is pressed against the transparent plate 11 while rotating. , the ground plane 61 is photographed by the camera 15. At this time, the tire 60 is photographed while being irradiated with light from a plurality of directions by a plurality of illumination lamps 16. Therefore, the camera 15 can photograph the tire 60 with a difference in brightness between the ground contact surface 61 and the portion other than the ground contact surface 61. The photographed image is acquired by the tire contact surface analysis device 20, and the tire contact surface analysis device 20 analyzes the contact surface 61 based on the acquired image.

Regarding the extraction of the groove portion of the tire 60, a distance sensor that detects differences in height (depth) using a triangulation method can also be used. However, in order to detect groove portions on the entire ground plane 61, it is necessary to scan the detection range of the distance sensor. Therefore, it is difficult to analyze the dynamic ground contact characteristics when the tire 60 is rolling just by using a distance sensor.

(Lighting conditions for shooting)
When acquiring an image of the contact area of the contact patch 61 of the tire 60, the illumination lamp 16 may be placed on the top surface 11U side of the transparent plate 11 so as to surround the contact patch 61. preferable. When acquiring an image of the contact area of the contact area 61 of the tire 60, it is preferable to acquire the image by irradiating light onto the tire 60 using an illumination lamp 16 arranged on the upper surface 11U side so as to surround the contact area. .

(Transparent plate movement and trigger device operation)
3 and 4 are diagrams for explaining the movement of the transparent plate 11 and the operation of the trigger device 17. FIG. 3 shows a state before the transparent plate 11 moves and a state before the trigger device 17 detects the light reflected by the retroreflective sheet 18. FIG. 4 shows the state after the transparent plate 11 has moved and the state when the trigger device 17 detects the reflected light by the retroreflective sheet 18.

In FIG. 3, the upper surface 11U of the transparent plate 11 is flat, and the upper surface 11U serves as a flat road surface on which the tires 60 roll. In FIG. 3, the tire 60 is fixed to the rim 4 of the support device 3 while being in contact with the upper surface 11U of the transparent plate 11. Therefore, as the transparent plate 11 moves, the tire 60 rotates. The tire ground contact shape analysis device 1 moves the transparent plate 11 in the direction of arrow Y1. By moving the transparent plate 11 in the direction of arrow Y1, the tire 60 rotates in the direction of arrow Y2. In the state shown in FIG. 3, the detection unit 171 of the trigger device 17 does not detect the light reflected by the retroreflection sheet 18. As long as the trigger device 17 does not detect the light reflected by the retroreflective sheet 18, the transparent plate 11 continues to move in the direction of the arrow Y1. Since the photographing device 10 is fixed to the transparent plate 11, the photographing device 10 also moves as the transparent plate 11 moves. The moving speed of the transparent plate 11 is, for example, 0.5 km/h. Note that instead of the transparent plate 11, a rotating drum having a transparent outer peripheral surface may be used.

When the transparent plate 11 moves in the direction of the arrow Y1 and enters the state shown in FIG. 4, the detection section 171 of the trigger device 17 detects the light reflected by the retroreflective sheet 18. When the detection unit 171 of the trigger device 17 detects the reflected light, the tire ground contact shape analysis device 1 outputs a photographing instruction signal to the camera 15. Thereby, the ground contact surface 61 of the tire 60 can be photographed. Note that the transparent plate 11 only needs to be transparent in a portion 110 corresponding to the photographing range of the camera 15, and may be opaque in portions other than the portion 110. In other words, the entire transparent plate 11 may be transparent, or only the portion 110 corresponding to the imaging range may be transparent.

(Operation of tire contact shape analysis device)
FIG. 5 is a flowchart showing the operation of the tire ground contact shape analysis device 1. When analyzing the tire 60, the tire ground contact shape analysis device 1 irradiates the tire 60 pressed against the transparent plate 11 with light from the illumination lamp 16 (step ST11). Next, the tire ground contact shape analysis device 1 starts driving the motor 6 by the motor control device 7 (step ST12). While continuing to drive the motor 6 (step ST13), the tire ground contact shape analysis device 1 determines whether the trigger device 17 has detected the retroreflective sheet 18 (step ST14). If the trigger device 17 does not detect the retroreflective sheet 18 (No in step ST14), the tire contact shape analysis device 1 continues to drive the motor 6 (step ST13).

When the trigger device 17 detects the retroreflective sheet 18 (Yes in step ST14), the tire ground contact shape analysis device 1 photographs the tire 60 with the camera 15 (step ST15). Thereafter, the tire ground contact shape analysis device 1 stops driving the motor 6 and stops emitting light (step ST16).

(Groove image extraction processing unit)
The groove image extraction processing unit 40 performs a process of extracting grooves using vertical lines. FIG. 6 is a diagram illustrating the process of extracting grooves using vertical lines. FIG. 7 is a flow diagram showing the process of extracting grooves using vertical lines.

In FIG. 6, first, a vertical line 71 is placed on the ground plane image SG acquired by photographing with the camera 15. The vertical line 71 is a normal to the boundary line of the temporary GCA image KG1.

In FIG. 7, first, a perpendicular line 71 to the boundary line of the provisional GCA image KG1 is set for the ground plane image SG (step ST21). A brightness profile along the vertical line 71 is extracted (step ST22). Groove extraction processing is performed using the brightness profile (step ST23). The groove extraction process will be described later.

The vertical line 71 is shifted in a direction parallel to itself (step ST24). In this example, the vertical line 71 is moved in the direction indicated by the arrow Y of the ground plane image SG1 in FIG. As a result of moving the vertical line 71, it is determined whether the vertical line 71 has reached the end of the ground plane image SG1 (step ST25). If it is determined in step ST25 that the vertical line 71 has not reached the end of the ground plane image SG1 (No in step ST25), the process returns to step ST22 and continues the process. Thereafter, the same process is repeated.

As a result of repeating similar processing, if the vertical line 71 reaches the end of the ground plane image SGn, it is determined in step ST25 that the vertical line 71 has reached the end of the ground plane image (step ST25 (Yes). In that case, next, it is determined whether or not all vertical lines 71 have been processed (step ST26). As a result of the determination in step ST26, if it is determined that all vertical lines 71 have not been processed (that is, there are vertical lines that have not been processed) (No in step ST26), the process returns to step ST22 and the process is continued. continue.

As a result of the determination in step ST26, if all the vertical lines 71 have been processed, the process ends (Yes in step ST26). Through the above processing, the groove image MG can be created from the ground plane image SG.

(Groove extraction processing)
The groove extraction process in step ST23 in FIG. 7 is performed. For the groove extraction process, for example, a method disclosed in Japanese Patent Application Laid-open No. 2017-129491 is adopted. An outline of the groove extraction process will be explained with reference to FIG. 8. FIG. 8 is a flow diagram showing the procedure of groove extraction processing. FIG. 9 is a diagram in which a vertical line 71 is added to the groove on the ground plane image SG. In FIG. 9, the pixel at the midpoint of the vertical line 71 is defined as the center pixel 74. FIG. 10 is a diagram showing an example of two maximum brightness positions PA and PB.

In FIG. 8, when performing the groove extraction process on the ground plane image SG, first, a vertical line 71, which is a brightness calculation line, is set (step ST31). A vertical line 71, which is a brightness calculation line, is a line for calculating the local brightness of a land area.

Next, it is determined whether there is a vertical line 71 protruding from the ground plane image SG (step ST32). As a result of this determination, if it is determined that there is a vertical line 71 protruding from the ground plane image SG (Yes in step ST32), the process exits from the flow of the groove extraction process.

If it is determined that there is no vertical line 71 protruding from the ground plane image SG (No in step ST32), the local brightness of the land area is calculated (step ST33). The land local brightness is calculated using the brightness information of pixels located in the brightness calculation area, which is the area where the vertical line 71 is located on the ground plane image SG and is the area consisting of the vertical line 71 and the neighboring area. .

Note that the land local brightness is the local brightness of the land area, so when calculating the land local brightness, it is necessary to exclude pixels corresponding to grooves on the ground plane image SG. is preferred. Specifically, since pixels corresponding to grooves have low luminance, by excluding pixels whose luminance is lower than a predetermined threshold value, pixels corresponding to grooves are excluded from calculation of average luminance.

Furthermore, since the brightness of the non-ground area is significantly different from the brightness of the ground area, when calculating the local brightness of land areas, it is preferable to exclude pixels corresponding to the non-ground area and calculate the average brightness. . Specifically, pixels corresponding to the non-ground area have higher luminance than pixels corresponding to the ground area, so by excluding pixels whose luminance is higher than a predetermined threshold, pixels corresponding to the non-ground area 82 are determined. , and exclude them from the calculation of average brightness.

After calculating the land local brightness, it is next determined whether the brightness of the center pixel 74 of the vertical line 71 is smaller than the groove local brightness (step ST34).

In the groove extraction process, the brightness of the center pixel 74 of the vertical line 71 is extracted, and the brightness of the center pixel 74 is compared with the groove local brightness. As a result of the comparison, if it is determined that the luminance of the center pixel 74 of the vertical line 71 is not smaller than the groove local luminance (No in step ST34), the flow exits from the groove extraction process.

On the other hand, if it is determined that the brightness of the center pixel 74 of the vertical line 71 is smaller than the groove local brightness (Yes in step ST34), the center pixel 74 is regarded as a component of the groove, and the center The local maximum brightness is extracted by scanning in both directions along the length of the vertical line 71 with the pixel 74 as the base point, and the maximum brightness position is determined (step ST35).

If it is determined that the brightness of the center pixel 74 is smaller than the groove local brightness, scanning is performed in a predetermined range from the center pixel 74 in the length direction of the vertical line 71 to extract the brightness, and then scanning is performed. The maximum brightness within the specified range is defined as the local maximum brightness. The local maximum brightness is extracted in both directions in the length direction of the scanning range starting from the center pixel 74. Further, in the length direction of the scanning range, the positions where pixels with local maximum brightness are located are determined as maximum brightness positions PA and PB. In FIG. 10, the vertical axis indicates brightness, and the horizontal axis indicates position [pixel] in the tire circumferential direction. The midpoint of the range of vertical line 71 is center pixel 74 .

Next, it is determined whether at least one of the local maximum brightnesses is greater than or equal to the land local brightness +α (step ST36). Here, the brightness of the pixel is smaller in the groove than in the land area, and the center pixel 74 is at a position where it is determined to be a component of the groove because the brightness is smaller than the groove local brightness. . Therefore, if the local maximum brightness when scanning the brightness in the length direction of the scanning range is higher than the local brightness of the land area, the maximum brightness positions PA and PB are located at the boundary between the groove and the land area. It can be estimated that

Note that α in this case may be set to a constant value in advance, or may be determined by examining the luminance distribution at a plurality of positions on the vertical line 71. In this embodiment, α=35 is set for the ground plane image 80 whose luminance is expressed in 256 gradations.

The land local brightness +α set in this way is compared with the local maximum brightness of each of the two maximum brightness positions PA and PB, and at least one local maximum brightness is determined as the land local brightness. It is determined whether the value is greater than or equal to +α. As a result of this determination, if it is determined that none of the local maximum brightnesses is greater than or equal to the land local brightness +α (No in step ST36), the process exits from the flow of the groove 65 extraction process.

On the other hand, if it is determined that at least one of the local maximum brightnesses is equal to or higher than the land local brightness +α (Yes in step ST36), then both of the local maximum brightnesses are It is determined whether or not the local brightness is greater than or equal to the local brightness +α (step ST37). As a result of this determination, if it is determined that the local maximum brightness at both locations is greater than or equal to the land local brightness +α (Yes at step ST37), the area surrounded by the two maximum brightness positions PA and PB is It is determined as a groove (step ST38).

On the other hand, if it is determined that the local maximum brightness is not greater than the land local brightness +α at both locations (No in step ST37), the area surrounded by one of the maximum brightness positions and the center pixel 74 is determined as a component of the groove 65 (step ST39).

(Temporary GCA calculation department)
FIG. 11 is a diagram illustrating the content of processing by the provisional GCA calculation unit 34. By smoothing the ground plane image SG obtained by photographing with the camera 15, a smoothed image HG is obtained. A binary image BG is obtained by binarizing the smoothed image using a uniform threshold value for each pixel. A temporary ground contact shape KS is obtained by subtracting the groove image MG from the binary image.

The groove image MG is the groove image MG in FIG. The contact surface image SG may be trimmed along the contour of the chamfer of the groove of the tire 60 (on the tread land side), and the groove may be extracted to obtain the groove image MG. A temporary GCA image KG1 is obtained by performing groove filling processing on the temporary ground contact shape KS. The groove filling process is a process of filling a portion corresponding to the groove of the tire 60 with black pixels.

(Lamp for lighting)
12 to 15 are diagrams illustrating the arrangement of each illumination lamp 16. FIG. 12 is a diagram showing the arrangement of each illumination lamp 16 from a direction along the rotation axis of the tire 60. As shown in FIG. 12, lamps 161 and 162 are provided on the upper surface 11U side, which is one main surface of the transparent plate 11. The lamps 161 and 162 may both be line illumination whose longitudinal direction is in the tire width direction. The lamps 161 and 162 are arranged at positions apart from each other in the tire circumferential direction of the tire 60. The lamp 161 and the lamp 162 are provided on different sides of the tire 60. That is, the lamps 161 and 162 are circumferential lamps that irradiate light onto the ground contact surface 61 from the outside to the inside in the tire circumferential direction. In this way, the lamps 161 and 162 are arranged so as to irradiate light onto the contact surface 61 of the tire 60 in the tire circumferential direction. The lamps 161 and 162 can increase the brightness of the chamfering of the sub-grooves.

Further, it is preferable that lamps 165, 166, 167, and 168 are provided on the lower surface 11D side, which is the other main surface of the transparent plate 11, as lamps on the other main surface side. The lamps 165, 166, 167, and 168 may all be line illumination whose longitudinal direction is the tire width direction. The lamps 167 and 168 are arranged at positions apart from each other in the tire circumferential direction of the tire 60. The lamp 167 and the lamp 168 are provided on different sides of the tire 60. That is, the lamps 167 and 168 are circumferential lamps that irradiate light onto the ground contact surface 61 from the outside to the inside in the tire circumferential direction. In this way, the lamps 167 and 168 are arranged so as to irradiate the contact surface 61 of the tire 60 with light in the tire circumferential direction. The lamps 167 and 168 can increase the brightness of the chamfering of the sub-grooves.

In FIG. 12, the heights from the centers of the light emitting surfaces of the lamps 161 and 162 to the upper surface 11U are designated H1 and H2. In FIG. 12, the inclination angles of the lamps 161 and 162, that is, the angles formed by the light irradiation direction with respect to the upper surface 11U of the transparent plate 11 are assumed to be θ1 and θ2. The angle θ1 is preferably 9.7°, and the angle θ2 is preferably 11.6°. In this embodiment, preferably the height H1 is 30 mm and the height H2 is 30 mm.

Further, in FIG. 12, the heights from the centers of the light emitting surfaces of the lamps 167 and 168 to the lower surface 11D are assumed to be H7 and H8. In FIG. 12, the inclination angles of the lamps 167 and 168, that is, the angles formed by the light irradiation direction with respect to the lower surface 11D of the transparent plate 11 are assumed to be θ7 and θ8. The angle θ7 is preferably 4.2°, and the angle θ8 is preferably 4.9°. In this embodiment, preferably the height H7 is 30 mm and the height H8 is 30 mm.

FIG. 13 is a diagram showing the arrangement of the illumination lamps 16 from the upper surface 11U side of the transparent plate 11. In FIG. 13, distance D1 is the distance from the center of the light emitting surface of lamp 161 to the center of the tire, distance D2 is the distance from the center of the light emitting surface of lamp 162 to the center of the tire, and distance D3 is the distance from the center of the light emitting surface of lamp 163 to the center of the tire. Distance, distance D4 is the distance from the center of the light emitting surface of the lamp 164 to the center of the tire, distance D5 is the distance from the center of the light emitting surface of the lamp 165 to the center of the tire, distance D6 is the distance from the center of the light emitting surface of the lamp 166 to the center of the tire, It is. Preferably, distance D1 = 260 mm, distance D2 = 260 mm, distance D3 = 230 mm, distance D4 = 230 mm, distance D5 = 215 mm, and distance D6 = 215 mm.

FIG. 14 is a view showing the arrangement of each illumination lamp 16 from a direction perpendicularly away from the rotation axis of the tire 60. The height H3 is the distance from the upper surface 11U of the transparent plate 11 to the center of the light emitting surface of the lamp 163, the height H4 is the distance from the upper surface 11U of the transparent plate 11 to the center of the light emitting surface of the lamp 164, and the height H5 is the distance of the transparent plate 11. The distance from the lower surface 11D to the center of the light emitting surface of the lamp 165, height H6, is the distance from the lower surface 11D of the transparent plate 11 to the center of the light emitting surface of the lamp 166. Let θ3 and θ4 be the inclination angles of the lamps 163 and 164, that is, the angles formed by the light irradiation direction with respect to the upper surface 11U of the transparent plate 11. The angle θ3 is preferably 0°, and the angle θ4 is preferably 0°. Let θ5 and θ6 be the inclination angles of the lamps 165 and 166, that is, the angles formed by the light irradiation direction with respect to the lower surface 11D of the transparent plate 11. The angle θ5 is preferably 21.8°, and the angle θ6 is preferably 20.1°. In this embodiment, the height H3 is preferably 13 mm, the height H4 is 13 mm, the height H5 is 62 mm, and the height H6 is 62 mm.

FIG. 15 is a diagram of the transparent plate 11 viewed from directly below the lower surface 11D side. In FIG. 15, distances in the tire circumferential direction from the center of the light emitting surface of each lamp 167, 168 to the center of the tire 60 are assumed to be D7, D8. In this embodiment, the distance D7 is preferably 183 mm, and the distance D8 is preferably 183 mm. It is preferable that the lamps 167 and 168 emit light parallel to the tire circumferential direction.

(smoothing process)
In this embodiment, the smoothed image HG in FIG. 11 is obtained by processing using a median filter on the ground plane image SG (hereinafter referred to as median processing). Median filter processing is a process of rearranging the pixel of interest and its surrounding pixels in descending order of luminance, and replacing the intermediate value with the luminance value of the pixel of interest. FIG. 16 is a diagram showing an example of a median filter. FIG. 16 shows calculation details for a 3×3 size including one pixel of interest (center pixel). By performing median filter processing, in this example, the luminance of the pixel of interest is replaced with the intermediate value "79". Smoothing by median filter processing is performed for the purpose of noise removal. Therefore, although the smoothing process is not an essential process, it is preferable to perform the smoothing process.

(shrinkage processing, expansion processing)
The groove filling process for obtaining the provisional GCA image KG1 in FIG. 11 will be described with reference to FIGS. 16 to 18.

FIG. 17 is an explanatory diagram of the contraction process. FIG. 18 is an explanatory diagram of the expansion process. For example, when the pixel of interest is a black pixel, the contraction process is a process of replacing the pixel of interest with a white pixel if there is even one white pixel around the pixel of interest, as shown in FIG. In other words, the contraction process takes each black pixel as the center pixel, and one of the eight surrounding pixels (one pixel each at the top left, top, top right, right, bottom right, bottom, bottom left, and left closest to the center pixel). However, if a white pixel exists, the center pixel is replaced with a white pixel. On the other hand, the expansion process is a process in which, as shown in FIG. 18, if there is even one black pixel around the pixel of interest, the pixel of interest is replaced with a black pixel. In other words, the dilation process uses each white pixel as the center pixel, and one of the eight surrounding pixels (one pixel each at the top left, top, top right, right, bottom right, bottom, bottom left, and left closest to the center pixel). However, if a black pixel exists, the central pixel is replaced with a black pixel.

In this embodiment, an area within a radius of Dd pixels from the center pixel is defined as a peripheral pixel. Then, Dd=1.5 is dilation processing, Dd=1.5 is deflation processing, Dd=1.5 is deflation processing, Dd=1.5 is dilation processing, Dd=2.5 is dilation processing, Dd=2 .5 for contraction processing, Dd = 2.5 for contraction processing, Dd = 2.5 for expansion processing, Dd = 3.5 for expansion processing, Dd = 3.5 for contraction processing, Dd = 3.5 for contraction processing , dilation processing at Dd=3.5, dilation processing at Dd=20.5, contraction processing at Dd=20.5, contraction processing at Dd=20.5, and expansion processing at Dd=20.5 are performed in this order. The processing up to this point is performed for the purpose of removing noise. Thereafter, in order to fill in the grooves and obtain a temporary GCA image, dilation processing is performed at Dd=110.5 and contraction processing is performed at Dd=110.5. Through the above processing, a temporary GCA image can be obtained.

(Vertical line installation processing section)
The vertical line installation processing unit 35 performs a process of installing a vertical line at each pixel on the boundary line for the temporary GCA image. FIG. 19 is a flowchart showing the processing contents of the vertical line installation processing section 35. FIG. 20 is a diagram illustrating a procedure for installing a vertical line by the process of the vertical line installation processing unit 35. FIG. 20 shows an enlarged portion of the temporary GCA image. In FIG. 20, the shaded areas are areas within the temporary GCA, and the unshaded areas are areas outside the temporary GCA. In FIG. 20, pixels on the boundary line are represented by rectangles.

In FIG. 19, first, an arbitrary pixel on the boundary line of the temporary GCA is defined as a center pixel (step ST41). Next, the length along the boundary line is defined as the sum of the distances between adjacent pixels using the center pixel as a base point (step ST42). Through the above processing, the sum of the distances shown by the reciprocating arrows in FIG. 20 becomes the length along the boundary line.

Returning to FIG. 19, the length along the boundary line is specified, and the pixels closest to the specified value are defined as peripheral pixels (step ST43). Two peripheral pixels sandwiching the center pixel are detected. Then, a straight line L0 passing through the center of each peripheral pixel is generated (step ST44). A vertical line 71 that passes through the center pixel and is perpendicular to the straight line L0 generated in step ST44 is generated in the temporary GCA (step ST45).

Next, it is determined whether all pixels on the boundary line of the temporary GCA have been processed (step ST46). As a result of the determination in step ST46, if it is determined that not all pixels have been processed (that is, there are pixels that have not been processed) (No in step ST46), the process returns to step ST41 and continues the process. As a result, similar processing is performed for other pixels on the boundary line of the temporary GCA.

As a result of the determination in step ST46, if all pixels have been processed (Yes in step ST46), the processing by the vertical line installation processing section 35 ends. Note that as a precondition for this process, it is assumed that there is no boundary between grounding and non-grounding outside the temporary GCA. Therefore, since no analysis is performed outside the tentative GCA, the vertical line VL is not displayed outside the tentative GCA.

(Luminance differential processing section)
Next, the luminance differentiation processing unit 36 differentiates the luminance distribution along each vertical line, and further differentiates the luminance distribution to find a maximum point and a minimum point. These processes will be explained below. FIG. 21 is a diagram illustrating the processing of the luminance differential processing section 36. FIG. 21 shows an enlarged portion of the temporary GCA image. In FIG. 21, the shaded areas are areas within the temporary GCA, and the unshaded areas are areas outside the temporary GCA. In FIG. 21, pixels on the boundary line are represented by rectangles.

(Advance preparation)
In preparation for analysis, perform the following processing. That is, for each plot point of a vertical line, the in-screen coordinates are defined as (X(N), Y(N)). In this example, "N" is set to a value in the range of -150 to +150. As shown in FIG. 21, N=-150 corresponds to the left end of the vertical line, and N=+150 corresponds to the right end of the vertical line. The angle θ, which is the inclination angle of the vertical line, is set with the X coordinate axis (positive direction) as a reference (θ=0°). The coordinates of the center pixel through which the vertical line passes are defined as (Xc, Yc).

In addition, in FIG. 21, the area from the angle θ=0° to θ=90° with respect to the center pixel as a reference is the first quadrant, and the area from the angle θ=90° to θ=180° is the second quadrant. The area from θ=180° to θ=270° is defined as the third quadrant, and the area from angle θ=270° to θ=0° is defined as the fourth quadrant.

Based on the above processing, perform the following processing.
(Step S11)
Calculate the in-screen coordinates X(N) and Y(N). If the angle θ is in the first or third quadrant,
X(N)=N×｜cosθ｜+Xc
Y(N)=-N×｜sinθ｜+Yc
It is. Also, if the angle θ is in the second or fourth quadrant,
X(N)=N×｜cosθ｜+Xc
Y(N)=N×｜sinθ｜+Yc
It is.

(Step S12)
Although the in-screen coordinates X(N) and Y(N) obtained in step S11 are real numbers, there is no brightness information. On the other hand, for digital images, brightness information is stored for each pixel. Therefore, using the brightness information of the digital image, the brightness of the in-screen coordinates X(N) and Y(N) is determined by linear interpolation.

FIG. 22 is a diagram illustrating the process of determining brightness by linear interpolation. In FIG. 22, the coordinates of the pixel positions of the digital image are (x, y), (x+1, y), (x+1, y+1), and (x, y+1). In FIG. 22, the direction from left to right is the positive direction of the X coordinate, and the direction from the top to the bottom is the positive direction of the Y coordinate.

The brightness of the coordinates (x, y) is "C1", the brightness of the coordinates (x+1, y) is "C2", the brightness of the coordinates (x+1, y+1) is "C3", the brightness of the coordinates (x, y+1) is "C4" ”. Brightness C1, C2, C3 and C4 are the brightness of the pixels of the digital image.

Assuming distance Xr=X(N)-x and distance Yr=Y(N)-y, the brightness D between coordinates (x, y) and coordinates (x+1, y) is calculated by linear interpolation.
Brightness D=C1×(1.0-Xr)+C2×Xr
can be asked.

Also, the brightness E between the coordinates (x, y+1) and the coordinates (x+1, y+1) is calculated by linear interpolation.
Brightness E=C4×(1.0-Xr)+C3×Xr
can be asked. Then, by linear interpolation using brightness D and brightness E,
Brightness F=D×(1.0-Yr)+E×Yr
can be asked. The brightness F is the brightness of the in-screen coordinates X(N), Y(N) determined by linear interpolation. The above calculation of brightness by linear interpolation is performed for all N values. In this example, the calculation is performed from N=-150 to N=+150.

By the way, there is noise such as uneven brightness and dirt on the ground surface. In order to remove the noise, it is preferable to smooth the ground plane image before performing step S12 above. FIG. 23 is a diagram illustrating the process of smoothing the ground plane image. FIG. 23 shows examples of images before and after smoothing processing. As shown in FIG. 23, the ground plane image SG in which the perpendicular line 71 to the temporary GCA boundary line KK is placed is smoothed by, for example, median processing. As a result, a smoothed ground plane image SG' is obtained.

(Step S13)
FIG. 24 is a diagram showing an example of brightness distribution. FIG. 24 is a graph plotting the brightness obtained in step S12 above. In FIG. 24, the left side is the outside of the tire, and the right side is the inside of the tire. The white circles in FIG. 24 are points where the brightness is plotted. Note that if the smoothing shown in FIG. 23 is not performed, the points plotted in step S13 may vary greatly, and the maximum and minimum points to be determined in the subsequent steps may not be reliably detected.

(Step S14)
Next, differentiate the plotted luminance. Here, with respect to the plot point of interest, the brightness of the plot points before and after it is quoted, and the difference is performed as a differential process. That is, the difference is regarded as a differential value. For example, when the plot point of interest is "50", the brightness D of the plot point "49" and the brightness E of the plot point "51" are quoted. In this embodiment, the difference is determined by subtracting the brightness D, which has a smaller absolute value, from the brightness E, which has a larger absolute value of the plot point. This difference is regarded as a differential value.

(Step S15)
Smoothing is performed to accurately detect local maximum points/minimum points. In this example, the brightness of the plot point of interest and the brightness of the plot points before and after it are used, and their arithmetic average is used for smoothing. Smoothing in step S13 is a process of finding an intermediate value of the collected luminance information. On the other hand, the smoothing in step S15 is a process in which the luminances are added together and divided by the number of data. The hatched circles in FIG. 24 are points where the luminance was differentiated in step S14 and further smoothed in step S15. Point P11 and point P13 in FIG. 24 are the maximum points, and point P12 is the minimum point.

(Step S16)
Further, differentiation processing is performed on the diagonally shaded circles in FIG. For example, when the plot point of interest is "50", the luminance differential value of plot point "49" and the luminance differential value of plot point "51" are quoted. In this example, the difference is determined by subtracting the smaller luminance differential value from the luminance differential value with the larger absolute value of the plot point. This difference is regarded as a second-order differential value. The black circles in FIG. 24 are points where the second-order differential values are plotted. By performing second-order differentiation, the point where the second-order differential value is maximum can be detected as the boundary point P14. Note that for all other vertical lines 71, steps S11 to S16 are performed after the above-mentioned advance preparation.

(Boundary setting process)
The boundary setting process by the boundary setting processing unit 37 will be explained. The boundary setting processing unit 37 sets boundaries in the provisional GCA image based on the maximum points and minimum points obtained by the luminance differentiation processing unit 36. The boundary setting processing unit 37 divides the cases based on the number of local maximum points in the plotted quadratic differential result of the brightness, and sets and determines boundary conditions according to the number of local maximum points, minimum points, and the like.

(If there are 0 local maximum points)
If the number of local maximum points is zero in the result of luminance differentiation, the boundary setting processing unit 37 does not set or determine the boundary conditions. In other words, if there are zero maximum points, no boundary is defined. This is because, in this case, the vertical line 71 never passes through the grounding block, but instead passes through the main groove.

(If there is one maximum point)
If there is one local maximum point in the brightness differential result, the boundary is defined as the position where the second-order differential is maximum among the plotted points from the first local maximum point position to the first local maximum point position +β. FIG. 25 is a diagram showing an example of a ground plane image. In this example, β=30. FIG. 26 is a diagram showing the luminance distribution along the vertical line 71A shown in FIG. 25.

The white circles in FIG. 26 are points where the brightness is plotted. The hatched circles in FIG. 26 are points where the luminance was differentiated and further smoothed. Point P21 in FIG. 26 is the maximum point. That is, there is one local maximum point. The black circles in FIG. 26 are points where the second-order differential values are plotted. By performing second-order differentiation, the point where the second-order differential value is maximum can be detected as the boundary point P24.

FIG. 27 is an erroneous determination map when the boundary determination process according to this example is performed on FIG. 26. As shown by ellipses J2 and J3 in FIG. 27, the error is small and is 11.6%. FIG. 28 is an erroneous determination map when the position of the first local maximum point in FIG. 26 is determined to be a boundary. As shown by ellipses J2' and J3' in FIG. 28, there is a larger error than in the case of FIG. 27, which is 12.1%. Therefore, when there is one maximum point, if the boundary is defined as described above, the error can be reduced. The boundary setting process regarding the vertical line 71A shown in FIG. 25 mainly affects the determination accuracy on the shoulder side of the tire, and has little effect on the determination accuracy on the center side of the tire.

(If there are two maximum points)
If there are two maximum points in the brightness differentiation result, there is one or more minimum points, and the plot points are arranged in the order of the first maximum point position, first minimum point position, and second maximum point position. In this case, the position where the second-order differential value is maximum among the plotted points from the position of the first minimum point to the position of the second maximum point is defined as the boundary. Otherwise, the position of the second maximum point is defined as the boundary.

FIG. 29 is a diagram showing the luminance distribution along the vertical line 71B shown in FIG. 25. The white circles in FIG. 29 are points where the brightness is plotted. The hatched circles in FIG. 29 are points where the luminance was differentiated and further smoothed. Points P31 and P33 in FIG. 29 are the maximum points, and point P32 is the minimum point. That is, there is one or more minimum points, and the plot points are arranged in the order of the first maximum point position, the first minimum point position, and the second maximum point position. Therefore, among the plotted points from the position of the first minimum point P32 to the position of the second maximum point P33, the point having the maximum second-order differential value can be detected as the boundary point P34.

FIG. 27 is an erroneous determination map when the boundary determination process according to this example is performed on FIG. 29. As shown by ellipses J4 and J5 in FIG. 27, the error is small and is 11.6%. FIG. 30 is an erroneous determination map when the position of the first minimum point in FIG. 29 is determined to be a boundary. As shown by the ellipse J4' in FIG. 30, there is a larger error than in the case of FIG. 27, which is 12.3%. FIG. 31 is an erroneous determination map when the position of the second local maximum point in FIG. 29 is determined to be a boundary. As shown by the ellipse J5' in FIG. 31, there is a larger error than in the case of FIG. 27, which is 11.8%.

Therefore, when there are two maximum points, if the boundary is defined as described above, the error can be reduced. The boundary setting process regarding the vertical line 71B shown in FIG. 25 affects the determination accuracy of the shoulder side and center side of the tire.

Here, FIG. 32 is an erroneous determination map when the boundary in FIG. 29 is determined based on other conditions. FIG. 32 shows an erroneous determination map when the plot point where the second-order differential is maximum is determined to be the boundary from the plot points from the position of the second maximum point to the position of the second maximum point +30 in FIG. 29. In this case, the error is small, 11.7%. For this reason, the plot point at which the second-order differential is the largest among the plot points from the position of the second maximum point to the position of the second maximum point +30 may be determined to be the boundary.

(If there are 3 or more maximum points)
FIG. 33 is a diagram showing the luminance distribution along the vertical line 71C shown in FIG. 25. The white circles in FIG. 33 are points where the brightness is plotted. The hatched circles in FIG. 33 are points where the luminance was differentiated and further smoothed. The hatched circle in FIG. 33 has many local maximum points and local minimum points, as well as a falling portion TD. If there are three or more maximum points as shown in FIG. 33 in the brightness differential results, the determination is made as follows.

(Step S31)
The position of the maximum point closest to the falling portion TD is searched from the plotted points from the position of the first maximum point to the position of the first maximum point +γ. However, for noise countermeasures, less than δ% of the luminance differential value at the first maximum point is excluded from the search. Note that in this example, γ=50 and δ=80.

A large number of local maximum points and local minimum points are considered to be caused by the influence of noise, and are treated as if they were one local maximum point. One maximum point is defined as the position of the maximum point closest to the falling part. In other words, the determination conditions are the same as in the case of one local maximum point, and the next step S32 is performed.

(Step S32)
Among the plotted points from the position of the maximum point closest to the falling portion TD detected in step S31 to the maximum point position +γ, the position where the second-order differential is maximum is defined as a boundary. Then, among the plotted points from the position of the maximum point P40 closest to the falling portion TD to the maximum point position +γ, the point having the maximum second-order differential value can be detected as the boundary point P44.

Returning to FIG. 27, FIG. 27 is an erroneous determination map when the boundary determination process according to this example is performed on FIG. 33. As shown by ellipses J3 and J6 in FIG. 27, the error is small, 11.6%. FIG. 34 is an erroneous determination map when boundaries are determined based on other conditions. FIG. 34 shows an erroneous determination map when the position of the local maximum point closest to the falling portion TD in FIG. 33 is determined to be the boundary. As shown by ellipses J3' and J6' in FIG. 34, there is a larger error than in the case of FIG. 27, which is 12.3%.

Therefore, when there are three or more maximum points, defining the boundaries as described above can reduce errors. The boundary setting process regarding the vertical line 71C shown in FIG. 25 mainly affects the determination accuracy on the shoulder side of the tire, and has little effect on the determination accuracy on the center side of the tire.

(Smoothing processing section)
Next, smoothing processing by the smoothing processing section 38 will be explained. The smoothing processing unit 38 performs smoothing processing to smooth the boundaries of the temporary GCA image whose boundaries have been set by the boundary setting processing unit 37. The smoothing processing unit 38 obtains the contour of the temporary GCA image in which the boundary has been set, and smoothes the contour.

FIG. 35 is a diagram illustrating the procedure for determining the contour. FIG. 35 shows an enlarged portion of the temporary GCA image. In FIG. 35, the shaded areas are areas within the temporary GCA, and the unshaded areas are areas outside the temporary GCA. In FIG. 35, pixels on the boundary line are represented by rectangles.

In FIG. 35, pixels forming the boundary line of the temporary GCA are labeled with numbers in a clockwise direction (in the direction of the arrow in the figure). The numbers inside each rectangle in FIG. 35 are labeling numbers. A vertical line 71 is provided for each pixel. In the ground/non-ground boundary position (X(m), Y(m)) calculated based on the luminance distribution, the argument m in parentheses is made to match the labeling number. FIG. 35 shows only vertical lines 71 passing through (X(1), Y(1)) and (X(14), Y(14)), respectively. FIG. 36 is a diagram showing a state in which vertical lines 71 are set for each pixel.

Note that vertical lines whose boundary positions (X(m), Y(m)) are not determined are also included. As described above, if the number of local maximum points is "0", the boundary position is not determined. In this example, the boundary position is not determined for the pixel with the labeling number "2". Therefore, in this example, the vertical line 71 is not displayed for (X(2), Y(2)) as an example in which the boundary position cannot be determined. Therefore, when creating a contour, only the determined boundary positions are used, and the arguments in parentheses are connected in order (that is, in ascending order) by a straight line SS. For example, if there is m=2 whose boundary position has not been determined, the boundary position at m=1 and the boundary position at m=3 are connected by a straight line SS.

(contour smoothing)
In contact surface images, the boundary position between contact and non-contact is often unstable due to uneven brightness and dirt on the tread surface. In order to stabilize it, it is preferable to smooth the contour. In order to smooth the contour, for example, the following process is performed.

(Step S21)
FIG. 37 is a diagram showing an example of the coordinates of the boundary position. FIG. 37 shows an example in which the coordinates of the determined boundary positions are arranged in the order of the labeling numbers. Boundary positions that have not been determined do not exist in FIG. For example, old numbers "2" and "6" are excluded because their boundary positions have not been determined, and corresponding new numbers do not exist for these.

(Step S22)
The boundary positions are smoothed by arithmetic averaging. Specifically, the boundary positions before and after the new number of interest are taken in and arithmetic averaged. For example, if the new number you are looking at is "3", the two numbers before and after it, that is, the boundary positions of the new numbers "1", "2", "4", and "5", are imported, and the new number "3" is obtained. A total of five boundary positions including the above are arithmetic averaged.

(Boundary judgment results)
FIGS. 38 and 39 are diagrams illustrating the boundary determination results before and after contour smoothing. FIG. 38 is a diagram illustrating an example of a boundary determination result before contour smoothing. FIG. 39 is a diagram showing an example of a boundary determination result after contour smoothing.

In FIG. 38, the erroneous determination map GM1 for the image RG1 before the contour smoothing of the boundary determination result has an unstable boundary, for example, as shown by an ellipse J1. Therefore, the error increases for image RG1. In this example, the error is 13.5% for the image RG1 before contour smoothing.

The misjudgment map is an image in which areas where the ground contact determinations do not match regarding the ground contact shape of the inked footprint are visualized in black. An area where the ground contact determination does not match is an area where the ground contact shape obtained by photographing with the camera 15 and the ground contact shape of the inked footprint do not match. Inked footprints are footprints that can be obtained by pressing a tire coated with ink against the tread. The error obtained from the erroneous determination map is the percentage of the area of the region where the contact determinations do not match divided by the actual contact area (hereinafter abbreviated as ACA) of the contact shape of the inked footprint.

In the erroneous determination map GM2 for the image RG2 after contour smoothing of the boundary determination result shown in FIG. 39, the boundary is stable with respect to the image RG1 before contour smoothing shown in FIG. Therefore, the error in the image RG2 after contour smoothing is 11.6%. Therefore, it is preferable to smooth the contour.

(Overlay processing section)
Next, the superposition processing by the superposition processing section 39 will be explained. FIG. 40 is a diagram illustrating the overlapping process. In FIG. 40, image RG2 is a temporary GCA image whose boundaries, that is, contours, have been smoothed by the smoothing processing unit 38. The superimposition processing unit 39
This contour-smoothed image RG2 and the groove image MG are superimposed. The superposition process by the superposition processing unit 39 is a process of inverting the black and white of the groove image MG and subtracting it from the image RG2. Through this superposition process, a ground contact shape image SK is obtained.

(length of vertical line)
The length of the vertical line 71 in the vertical line installation processing unit 35 that installs the vertical line 71 with respect to the boundary line of the temporary GCA will be explained. FIG. 41 is a diagram illustrating the length of the vertical line set by the vertical line installation processing unit 35. In FIG. 41, the length L71 of the vertical line 71 is defined as the length in the direction toward the inside of the tire from the starting point SP on the temporary GCA boundary line KK (gray line in FIG. 41). For example, the length L71 is preferably 1% or more and 5% or less of the tire outer diameter, and more preferably 1.5% or more and 3.8% or less. By setting the length L71 of the vertical line 71 within such a numerical range, the accuracy of determining the boundary between ground contact and non-ground contact is further improved. If the length of the vertical line is less than 1% of the tire outer diameter, there is insufficient information regarding the maximum/minimum points of the brightness differential, and the accuracy of boundary determination decreases. When the length of the vertical line 71 exceeds 5% of the tire outer diameter, the vertical line reaches the chamfered portion of the main groove or sub-groove in the ground contact image. Since the chamfer has high brightness, a maximum point exists. For this reason, the accuracy of boundary determination decreases, which is not preferable.

(Threshold value of provisional GCA calculation unit)
In the provisional GCA calculation unit 34, it is preferable that the threshold value when the ground plane image is binarized is smaller than the maximum brightness value in the ground plane image. By appropriately setting the threshold value when binarizing the ground contact surface image, it is possible to improve the accuracy of the ground contact/non-ground contact boundary determination.

Here, in this example, the ground plane image has 256 gradations, the minimum brightness value is "0", and the maximum brightness value is "255". Note that the threshold value may be the minimum brightness "0". This is because there is a tire contact surface image in which the brightness of the contact block is 0.

42 and 43 are diagrams showing examples of the results of binarizing the ground plane image in the temporary GCA calculation unit 34. FIG. 42 is a diagram showing an example in which the threshold is set to 220 gradations, which is a threshold smaller than the maximum brightness value "255". In FIG. 42, the black part is the temporary GCA image KG1, and the gray part is the ground contact shape of the inked footprint. In FIG. 42, the boundary position of the ground contact shape of the inked footprint is included in the temporary GCA image KG1. Therefore, it is possible to ensure accuracy in determining the boundary between grounding and non-grounding.

FIG. 43 is a diagram showing an example in which the threshold is set to 254 gradations, which is one threshold smaller than the maximum brightness value "255". In FIG. 43, the black part is the provisional GCA image KG1, and the gray part is the ground contact shape of the inked footprint. In FIG. 43, the boundary position of the ground contact shape of the inked footprint is included in the temporary GCA image KG1. Therefore, it is possible to ensure accuracy in determining the boundary between grounding and non-grounding.

Note that when the threshold value is set to the maximum brightness value "255", all non-grounded blocks are taken in and formed as a temporary GCA. Therefore, the entire area of the temporary GCA image is completely black. Due to this influence, the groove in the non-ground contact area is mistakenly determined as a boundary, and the accuracy of determining the contact/non-contact boundary is reduced. Therefore, it is inappropriate as a total grounding area. Therefore, as described above, the threshold value when binarizing the ground plane image is preferably smaller than the maximum brightness value in the ground plane image.

(summary)
According to the tire contact shape analysis device described above, the boundary condition of contact or non-contact can be set based on the maximum point or minimum point of the plotted second-order differential value of brightness, and the contact shape can be appropriately determined. .

(Tire contact shape analysis method)
The tire contact shape analysis device described above can implement the following tire contact shape analysis method. That is, there is an image acquisition step for acquiring a tire contact surface image, and a binarization process is performed on the contact surface image acquired in the image acquisition step using a predetermined threshold as a reference to obtain a provisional total contact area image. A provisional GCA calculation step for obtaining a GCA image, a vertical line installation step for installing a vertical line at each pixel on the boundary line for the provisional GCA image, and a vertical line installation step for each vertical line installed in the vertical line installation processing step. A brightness differentiation processing step in which the distribution of brightness along the line is differentiated and further differentiated to obtain maximum points and minimum points, and the provisional GCA image is calculated based on the maximum points and minimum points obtained by the brightness differentiation processing step. a boundary setting processing step for setting a boundary in the boundary setting processing step; a smoothing processing step for smoothing the boundary for the temporary GCA image in which the boundary has been set by the boundary setting processing step; and a smoothing processing step for smoothing the boundary by the smoothing processing step. It is possible to realize a tire ground contact shape analysis method including a superposition processing step of superimposing a groove image obtained from the ground contact surface image on the temporary GCA image obtained by the above process to obtain a ground contact shape. According to this tire contact shape analysis method, it is possible to appropriately determine the contact shape by setting the boundary condition of contact or non-contact based on the number of maximum or minimum points of the plotted second-order differential value of brightness. .

The present disclosure includes the following inventions.
Invention [1]
an image acquisition unit that acquires a tire contact surface image;
a temporary GCA calculation unit that performs a binarization process on the ground contact surface image acquired by the image acquisition unit based on a predetermined threshold value to obtain a temporary GCA image that is a provisional total ground contact area image;
a vertical line installation processing unit that installs a vertical line at each pixel on the boundary line for the temporary GCA image;
a luminance differentiation processing unit that differentiates the distribution of luminance along the vertical line for each vertical line installed by the vertical line installation processing unit, and further differentiates the luminance distribution to obtain a local maximum point and a local minimum point;
a boundary setting processing unit that sets a boundary in the provisional GCA image based on the maximum point and minimum point obtained by the luminance differential processing unit;
a smoothing processing unit that smooths the boundaries of the temporary GCA image in which boundaries have been set by the boundary setting processing unit;
a superimposition processing unit that obtains a ground contact shape by superimposing a groove image obtained from the ground contact surface image on the provisional GCA image whose boundary has been smoothed by the smoothing processing unit;
A tire contact shape analysis device including:
Invention [2]
The tire ground contact shape analysis device according to invention [1], further comprising a groove image extraction processing unit that obtains the groove image based on the ground contact surface image.
Invention [3]
The length of the vertical line installed in the vertical line installation processing unit is defined as the length in the direction toward the inside of the tire starting from the boundary line of the temporary GCA image, and the length is 1% or more of the tire outer diameter. The tire ground contact shape analysis device according to invention [1] or invention [2], wherein the tire contact shape analysis device is 5% or less.
invention [4]
The tire ground contact shape analysis device according to any one of inventions [1] to [3], wherein in the binarization process of the provisional GCA calculation unit, the predetermined threshold value is smaller than the maximum brightness value in the ground contact surface image. .
invention [5]
The boundary setting processing unit determines the maximum and minimum points of brightness plotted on the graph.
When there is one local maximum point, the position where the second-order differential value is maximum among the plotted points from the position of the first local maximum point to the position of the first local maximum point + β is defined as the boundary,
When there are two maximum points, if there is one or more minimum points and the plot points are arranged in the order of the first maximum point, first minimum point, and second maximum point, then Among the plotted points from the position of the first local minimum point to the position of the second local maximum point, the position where the second-order differential value is maximum is defined as the boundary, and if there are two local maximum points and other than the above, the second local maximum Define the position of the point as the boundary,
If there are three or more maximum points, the boundary is defined as the position where the second derivative value is maximum among the plotted points from the position of the maximum point closest to the falling part of the graph to the position of the maximum point + γ. ,
The tire contact shape analysis device according to any one of inventions [1] to [4], which does not define boundaries when the number of local maximum points is zero.
invention [6]
an image acquisition step of acquiring a tire contact surface image;
A provisional GCA calculation step of performing binarization processing on the ground contact surface image obtained in the image acquisition step using a predetermined threshold as a reference to obtain a provisional GCA image that is a provisional total ground contact area image;
a vertical line setting processing step of setting a vertical line at each pixel on the boundary line for the temporary GCA image;
For each vertical line installed in the vertical line installation processing step, a brightness differentiation processing step of differentiating the luminance distribution along the vertical line and further differentiating to obtain a maximum point and a minimum point;
a boundary setting process step of setting a boundary in the provisional GCA image based on the maximum point and minimum point obtained in the brightness differentiation process step;
a smoothing process step of smoothing the boundary of the temporary GCA image in which the boundary has been set in the boundary setting process step;
a superposition processing step of superimposing a groove image obtained from the ground contact surface image on the provisional GCA image whose boundary has been smoothed by the smoothing processing step to obtain a ground contact shape;
A tire contact shape analysis method including

1 Tire contact shape analysis device 2 Tire testing machine 3 Support device 4 Rim 5 Drive device 6 Motor 7 Motor control device 10 Photographing device 11 Transparent plate 15 Camera 16 Illumination lamp 17 Trigger device 18 Retroreflective sheet 20 Tire contact surface analysis device 21 Input section 22 Display section 30 Processing device 31 Processing section 32 Storage section 33 Image acquisition section 34 Temporary GCA calculation section 35 Vertical line installation processing section 36 Luminance differential processing section 37 Boundary setting processing section 38 Smoothing processing section 39 Overlay processing section 40 Groove image extraction processing unit 60 Tire 61 Ground contact surface 71, 71A, 71B, 71C Vertical line

Claims (6)

an image acquisition unit that acquires a tire contact surface image;
a temporary GCA calculation unit that performs a binarization process on the ground contact surface image acquired by the image acquisition unit based on a predetermined threshold value to obtain a temporary GCA image that is a provisional total ground contact area image;
a vertical line installation processing unit that installs a vertical line at each pixel on the boundary line for the temporary GCA image;
a luminance differentiation processing unit that differentiates the distribution of luminance along the vertical line for each vertical line installed by the vertical line installation processing unit, and further differentiates the luminance distribution to obtain a local maximum point and a local minimum point;
a boundary setting processing unit that sets a boundary in the provisional GCA image based on the maximum point and minimum point obtained by the luminance differential processing unit;
a smoothing processing unit that smooths the boundaries of the temporary GCA image in which boundaries have been set by the boundary setting processing unit;
a superimposition processing unit that obtains a ground contact shape by superimposing a groove image obtained from the ground contact surface image on the provisional GCA image whose boundary has been smoothed by the smoothing processing unit;
A tire contact shape analysis device including: The tire contact shape analysis device according to claim 1, further comprising a groove image extraction processing unit that obtains the groove image based on the contact surface image. The length of the vertical line installed in the vertical line installation processing unit is defined as the length in the direction toward the inside of the tire starting from the boundary line of the temporary GCA image, and the length is 1% or more of the tire outer diameter. The tire ground contact shape analysis device according to claim 1 or 2, wherein the tire ground contact shape analysis device is 5% or less. The tire ground contact shape analysis device according to claim 1 or 2, wherein in the binarization process of the provisional GCA calculation unit, the predetermined threshold value is smaller than the maximum brightness value in the ground contact surface image. The boundary setting processing unit determines the maximum and minimum points of luminance plotted on the graph.
When there is one local maximum point, the position where the second-order differential value is maximum among the plotted points from the position of the first local maximum point to the position of the first local maximum point + β is defined as the boundary,
When there are two maximum points, if there is one or more minimum points and the plot points are arranged in the order of the first maximum point, first minimum point, and second maximum point, then Among the plotted points from the position of the first local minimum point to the position of the second local maximum point, the position where the second-order differential value is maximum is defined as the boundary, and if there are two local maximum points and other than the above, the second local maximum Define the position of the point as the boundary,
If there are three or more maximum points, the boundary is defined as the position where the second derivative value is maximum among the plotted points from the position of the maximum point closest to the falling part of the graph to the position of the maximum point + γ. ,
If there are 0 maximum points, no boundary is defined.
The tire ground contact shape analysis device according to claim 1 or 2. an image acquisition step of acquiring a tire contact surface image;
A provisional GCA calculation step of performing binarization processing on the ground contact surface image obtained in the image acquisition step using a predetermined threshold as a reference to obtain a provisional GCA image that is a provisional total ground contact area image;
a vertical line installation processing step of installing a vertical line at each pixel on the boundary line for the temporary GCA image;
For each vertical line installed in the vertical line installation processing step, a brightness differentiation processing step of differentiating the luminance distribution along the vertical line and further differentiating to obtain a maximum point and a minimum point;
a boundary setting process step of setting a boundary in the provisional GCA image based on the maximum point and minimum point obtained in the brightness differentiation process step;
a smoothing process step of smoothing the boundary of the temporary GCA image in which the boundary has been set in the boundary setting process step;
a superposition processing step of superimposing a groove image obtained from the ground contact surface image on the provisional GCA image whose boundary has been smoothed by the smoothing processing step to obtain a ground contact shape;
Tire contact shape analysis method including

Images