Method and system for automatically detecting nail inlaying position of tread of antiskid tire

Abstract

The invention discloses a method and a system for automatically detecting the nailing position of an anti-skid tire tread. The method comprises the following steps: acquiring a brightness map and point cloud data of the anti-skid tire tread; processing the point cloud data to generate a height map and an X value map; carrying out threshold segmentation on the brightness image to generate binary images, carrying out image sharpening processing on the height image, carrying out threshold segmentation to generate binary images, comparing each pixel point of the two binary images, generating an image containing a minimum value, and carrying out inversion and morphological processing on the pixel of the minimum value image; performing connected domain analysis on the processed image, screening connected domains by utilizing the characteristic difference between the nailing position and other areas, and displaying a screening result; and calculating the center point coordinates of the screened connected domains in the image, then calculating the three-dimensional coordinates of the center of the nailing position, and storing the position data in a file with a required format. According to the invention, the nail inlaying position of the antiskid tire tread can be rapidly and accurately detected, and the subsequent antiskid tire manufacturing or refitting process is facilitated.

Description

Technical Field

[0001] The invention relates to an automatic detection method and system for the position of nails embedded in the tread of an anti-skid tire. Background Art

[0002] When a vehicle is driving on a snowy or slippery road, it is easy to slip, thus affecting the normal driving of the vehicle. In order to solve the slipping problem, the method of fastening anti-skid chains on the tires is usually adopted. This method is inconvenient to use and increases the weight of the tires, seriously affecting the driving speed of the vehicle. The best solution at present is to embed metal nails or other anti-skid elements in the tire tread.

[0003] Traditional anti-skid tire studding processes usually rely on manual or semi-automatic equipment. Workers locate the studs on the tread by visual inspection, which is inefficient and not very accurate. In addition, some existing automated equipment relies on simple geometric models or preset templates to set the stud positions. Although this method can improve efficiency to a certain extent, it is still difficult to achieve precise positioning on complex tire patterns or irregular treads. Summary of the invention

[0004] The present invention aims to solve the problems existing in the prior art and provides an automatic detection method and system for the position of studs on the tread of an anti-skid tire.

[0005] The technical solutions adopted in the present invention are:

[0006] An automatic detection method for the position of studs on the tread of an anti-skid tire comprises the following steps:

[0007] S1: Obtain the brightness image and point cloud data of the anti-skid tire tread, and perform threshold segmentation on the brightness image to generate a binary image;

[0008] S2: Process the point cloud data to generate a height map and an X-value map, perform image sharpening on the height map, and then perform threshold segmentation to generate a binary image;

[0009] S3: Compare the pixel values ​​of each pixel of the two binary images S1 and S2, retain the pixel with the smaller pixel value, and finally generate an image with the minimum pixel value, and then invert the pixels of the image with the minimum pixel value;

[0010] S4: Perform morphological processing on the minimum pixel value image after inversion in S3;

[0011] S5: Performing connected domain analysis on the image after morphological processing, screening connected domains by using feature differences between the pinning position and other regions in the connected domain that are not pinning regions, and displaying the screening results;

[0012] S6: Calculate the coordinates of the center point of the connected domain after screening in the image, then calculate the three-dimensional coordinates of the center of the nail position, and save the position data in a required format file.

[0013] Furthermore, the point cloud data is processed to generate a height map and an X-value map, specifically:

[0014] Extract the X, Y, and Z coordinates of each point in the point cloud data and define the grid resolution, whose size is determined by the brightness map resolution;

[0015] The space is divided into grids, and the Z coordinate value and X coordinate value of each grid point in the point cloud are counted. The corresponding Z coordinate value and X coordinate value average of all points in each grid are taken, and the corresponding average values ​​of all grids are mapped to the two-dimensional image to generate the height map and X value map accordingly.

[0016] Furthermore, for grids with no points, the value is set to -327.

[0017] Furthermore, the height map image is sharpened by defining a convolution kernel, and the parameters of the convolution kernel are set using the binary image of the generated brightness map.

[0018] Furthermore, the parameter value in the convolution kernel is adjusted according to the brightness of the sliding position in the height map. Figure 2 The corresponding pixel position in the value image is determined by:

[0019] When the brightness Figure 2 When the pixel at this point in the value image is 0, the value of the corresponding point of the convolution kernel is 0;

[0020] When the brightness Figure 2 When the pixel value of this point in the value image is 255, the value of this point in the convolution kernel is -1;

[0021] If the midpoint of the convolution kernel exceeds the image range, the value of the out-of-range point is set to 0;

[0022] The value of the center point of the convolution kernel is the number of points with a value of -1 in the convolution kernel.

[0023] Furthermore, the defined convolution kernel is used to traverse each pixel of the height map.

[0024] If the pixel value of the corresponding point in the binary image of the brightness map is 0, the value of the point in the height map is replaced by 0;

[0025] If the brightness Figure 2 If the pixel in the center of the value image is 255, the value of the point in the height map replaces the value of the pixel in the original height image with the calculated value. The calculation method is: the corresponding pixel point and the neighboring pixel are multiplied and accumulated with the convolution kernel and divided by the value of the center point of the convolution kernel plus 1;

[0026] The pixel calculation formula for sharpening the height map is:

[0027]

[0028] in:

[0029] The height map is H (x，y) , the binary image of the brightness image is I (x，y) , the convolution kernel is F (i，j) , whose size is (2m+1)×(2n+1), F (0，0) is the value of the center point of the convolution kernel corresponding to the calculated pixel.

[0030] Furthermore, the morphological processing is:

[0031] First, the minimum image is eroded using a 3×3 rectangular structure element, then the minimum image is expanded using a 5×5 rectangular structure element, and finally the minimum image is eroded again using a 3×3 rectangular structure element.

[0032] Furthermore, the connected domain analysis is specifically as follows:

[0033] S51) according to the centroid coordinates of each connected domain, filter and remove the connected domains whose pixel distance between the centroid and the image boundary is less than the length D;

[0034] S52) calculating the area Area of ​​each connected domain, that is, the number of pixels contained in the connected domain, and filtering out the connected domains whose area is smaller than the minimum area threshold and whose area is larger than the maximum area threshold;

[0035] S53) calculating the structural feature SF=A×B-1 of the connected domain, and filtering out the connected domains whose feature values ​​exceed 1;

[0036] Where: A=Ra / R b ; B = (π × R a ×R b ) / Area;

[0037] A is the ratio of the major axis to the minor axis of each connected domain; B is the ratio of the product of the major axis, the minor axis and the pi of each connected domain to the area;

[0038] S54) Calculate the inclination angles of each connected domain in the X and Y directions, select four regions with a length of w pixels and a width of h pixels in the left and right directions and the up and down directions of d pixels away from the centroid of each connected domain, calculate the mean values ​​of the regional pixels in the height map and the X value map, which are Z1, Z2, Z3, Z4 and X1, X2 respectively, and the calculation formula of the inclination angle of the connected domain is: Angel x =atan2(Z2-Z1,X2-X1),Angel y= atan2(Z4-Z3,2d×t);

[0039] Among them, atan2 is the inverse tangent value, t is the resolution of the image in the Y direction, and the images with an inclination angle greater than the threshold θ in the X or Y direction are removed. x Or θ y connected domain.

[0040] Furthermore, the three-dimensional coordinate calculation process of the center of the nailing position is:

[0041] The remaining area after screening is the studded area in the anti-skid tire. The center point coordinates of the studded area in the anti-skid tire are obtained by the centroid coordinates of the screened connected domain. The calculation formula is: (x, y) = (x t +0.5,y t +0.5), where x t is the horizontal coordinate of the centroid of the connected domain, y t is the ordinate of the centroid of the connected domain;

[0042] The three-dimensional coordinates of the center point are calculated according to the coordinates of the center point of the pinning position area in the minimum pixel value image, and the pixel average calculated for the selected area around each pinning position in the height map and the X value map. The calculation formula is: Y = y × t,

[0043] The present invention also provides an automatic detection system for the position of studs on the tread of an anti-skid tire, comprising:

[0044] A collection module, the collection module is used to collect brightness images and point cloud data of the anti-skid tire tread;

[0045] A point cloud processing module, which is used to process point cloud data and generate a height map and an X-value map;

[0046] An image processing module, wherein the image processing module is used to segment the brightness image and the height image to generate binary images, and then compare each pixel point of the binary image of the brightness image and the binary image of the height image to generate an image containing a minimum value, and then invert the pixels of the minimum pixel value image and perform morphological processing and connected domain analysis, and use the feature differences of other areas in the connected domain that are not the studded area to screen the connected domain, calculate the coordinates of the center point of the screened connected domain in the image, and calculate the three-dimensional coordinates of the center of the studded position;

[0047] A display module, which is used to visualize the processing results and show the detection results of the nailing position to the user;

[0048] The saving module is used to save the nailing position coordinate data obtained by detection in a required format and record the operation log of the system.

[0049] The present invention has the following beneficial effects:

[0050] The present invention collects the brightness map and point cloud data of the tire tread, and processes the point cloud data to generate a height map and an X-value map, and processes the two-dimensional image to accelerate the detection efficiency. Considering that a large number of noise areas will be retained if the height image or the brightness image is directly threshold segmented, which is not conducive to subsequent screening, the present invention uses the binary image of the brightness image to set the convolution kernel parameters, sharpen the height map, and then threshold segment it, and merge the two binary images, take the smaller value of each pixel point of the two, and generate a new binary image, thereby removing a large number of interference areas. Considering that there will be more texture grooves on the anti-skid tire, the angle characteristics of the regional inclination are considered, and the connected domain is screened to improve the detection accuracy. Considering that the studded position of the anti-skid tire is a hole area, the camera cannot directly collect the data of the middle area, and the three-dimensional coordinates of the center of the studded position are indirectly calculated through the surrounding area. Finally, the results are displayed, and the studded position data is saved in the required format file, and the operation log of the recording system is recorded, which is convenient for the subsequent anti-skid tire manufacturing or modification process. The overall detection process time of the present invention is about 15s, which meets the production needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention is a schematic diagram of steps of an automatic detection method for the position of studs on the tread of an anti-skid tire.

[0052] Figure 2 It is a schematic diagram of the collection device.

[0053] Figure 3(a) is a height map generated from the anti-skid tire point cloud data.

[0054] Figure 3(b) is the X-value graph generated from the anti-skid tire point cloud data.

[0055] Figure 4 It is a schematic diagram of the selected area around the connected domain in the height map.

[0056] Figure 5 This is a schematic diagram showing the screening results in a brightness graph. DETAILED DESCRIPTION

[0057] The present invention will be further described below in conjunction with the accompanying drawings.

[0058] like Figure 1 The present invention provides an automatic detection method for the position of studs on the tread of an anti-skid tire, comprising:

[0059] Step S1: obtaining a brightness image and point cloud data of the anti-skid tire tread;

[0060] By starting the acquisition device, the brightness image and point cloud data of the tread of the anti-skid tire can be obtained through the 3D camera 2 fixed on the tread of the anti-skid tire 1 with a line scanning laser.

[0061] like Figure 2 As shown, the acquisition device of the present invention includes a bracket 3, a motor 4 and a roller 5. The motor 4 and the 3D camera 2 are both installed on the bracket 3. A rotating shaft 6 is rotatably connected to the bracket 3. The anti-skid tire 1 is fixed to the rotating shaft 6. The roller 5 is fixed to the motor shaft of the motor 4. The roller 5 fits the inner wall surface of the anti-skid tire 1. When the start button is pressed, the motor 4 rotates at a preset speed, and the roller 5 connected to the motor also rotates accordingly. The friction between the roller and the inner wall of the tire causes the tire to rotate together.

[0062] Step S2: Processing the point cloud data to generate a height map and an X-value map;

[0063] Extract the X, Y, and Z coordinates of each point in the point cloud. Define the grid resolution, whose size is determined by the brightness map resolution, divide the space into grids, and for each grid point, count the Z and X values ​​of the point cloud it contains, and take the average value of all points in each grid. For grids without points, set their values ​​to -327.

[0064] The processed Z value, i.e., the height, is mapped onto the two-dimensional image to generate a height map. Similarly, the processed X value is mapped onto the two-dimensional image to generate an X value map, as shown in FIG3(a) and FIG3(b).

[0065] Step S3: Threshold segmentation is performed on the brightness image to generate a binary image, and then the height image is sharpened and then threshold segmentation is performed to generate a binary image. Then, the brightness image and the height binary image are compared, and each pixel of the two binary images is compared to generate an image containing the minimum value, and then the pixels of the minimum value image are inverted.

[0066] The present invention sets a suitable threshold value first, and performs threshold segmentation on the brightness image to generate a binary image.

[0067] The present invention performs sharpening processing on the height map image by defining a convolution kernel, and uses the binary image of the generated brightness map to set the parameters of the convolution kernel. Specifically: the parameter value in the convolution kernel is based on the sliding position in the height map in the brightness Figure 2 The corresponding pixel position in the value image is determined.

[0068] When the brightness Figure 2 When the pixel at this point in the value image is 0, the value of the corresponding point of the convolution kernel is 0;

[0069] When the brightness Figure 2 When the pixel value of this point in the value image is 255, the value of this point in the convolution kernel is -1;

[0070] If the midpoint of the convolution kernel exceeds the image range, the value of this part of the point is also set to 0.

[0071] The value of the center point of the convolution kernel is the number of -1 values ​​in other points in the convolution kernel.

[0072] The binary image of the brightness map is I (x，y) , the convolution kernel is F (i，j) , its size is (2m+1)×(2n+1), the coordinates of the center point of the convolution kernel are (0,0), and the center point value is expressed as follows: Where i and j are not 0 at the same time, δ(θ) is:

[0073] The values ​​of points outside the center are expressed as follows:

[0074]

[0075] Use the set convolution kernel to slide on the height map, that is, traverse each pixel of the height map. If the pixel corresponding to the pixel in the binary image of the brightness map is 0, the value of the point in the height map is replaced by 0. Figure 2 The pixel in the midpoint of the value image is 255, then the value of the point in the height map is multiplied and accumulated for each pixel and the neighboring pixel with the convolution kernel and divided by the value of the convolution kernel center point plus 1, and the calculated value is used to replace the value of the pixel in the original image.

[0076] The height map is H (x,y) , the result of sharpening is H ( ′ x,y) , then the pixel calculation method for sharpening the height map can be expressed as follows:

[0077]

[0078] Then set a suitable threshold, perform threshold segmentation on the height map after sharpening, and generate a highly binary image. Then traverse each pixel of the image, compare the pixel size of each point of the binary image of the brightness image and the binary image of the height map after image filtering, and generate an image containing the minimum value. Then invert the pixels of the above result image. The pixels of the binary image are only 0 and 255. Pixel inversion is to change pixel 0 to pixel 255, and pixel 255 to pixel 0.

[0079] Step S4: performing morphological processing on the image;

[0080] First, the minimum image is eroded using a 3×3 rectangular structure element, then the minimum image is expanded using a 5×5 rectangular structure element, and finally the minimum image is eroded again using a 3×3 rectangular structure element. The erosion and expansion process can remove some noise areas in the image and merge some discrete areas into a larger area, which is convenient for the subsequent screening of connected domains.

[0081] Step S5: Performing a connected domain analysis on the processed image, using the feature differences between the nailing position and other regions (i.e., the non-nailed regions in the connected domain), screening the connected domain, and displaying the screening results;

[0082] The points of pixels adjacent to each other in the binary image processed in step S3 are placed in a region to generate the connected domain. The connected domain analysis is specifically as follows:

[0083] (1) First, based on the centroid coordinates of each connected domain, filter out the connected domains whose pixel distance between the centroid and the image boundary is less than the length D;

[0084] (2) According to the area characteristics of the connected domain, the area Area of ​​each connected domain, that is, the number of pixels contained in the connected domain, is calculated, and the connected domains with an area smaller than the minimum area threshold and an area larger than the maximum area threshold are screened out;

[0085] (3) According to the structural characteristics of the connected domain, the ratio of the major axis to the minor axis of each connected domain is calculated, that is, A = Ra / R b ; and the ratio of the product of the major axis, minor axis and pi to the area, that is, B = (π × R a ×R b ) / Area; finally, calculate the structural feature SF=A×B-1 of the connected domain; filter out the connected domains whose eigenvalues ​​exceed 1;

[0086] (4) According to the angle characteristics of the connected domain, calculate the inclination angles of each connected domain in the X and Y directions. Figure 4 As shown in the figure, four regions with a length of w pixels and a width of h pixels are selected in the left and right and up and down directions of d pixels from the centroid of each connected domain, and the mean values ​​of the regional pixels are calculated in the height map and the X value map, which are Z1, Z2, Z3, Z4 and X1, X2 respectively. The method for calculating the tilt angle of the connected domain is: Angel x =atan2(Z2-Z1,X2-X1),Angel y = atan2(Z4-Z3, 2d×t), where atan2 is the inverse tangent value of the two parameters, t is the resolution of the image in the Y direction, and the images with an inclination angle greater than the threshold θ in the X or Y direction are removed. x Or θ y connected domain.

[0087] like Figure 5 As shown, the filtered nail position area is displayed in the brightness map to show the detection result of the nail position to the user.

[0088] Step S6: Calculate the coordinates of the center point of the connected domain after screening in the image, then calculate the three-dimensional coordinates of the center of the nail position, and save the coordinate data in a required format file.

[0089] Through the above screening of connected domains, the connected domains retained are the nailing position areas of the anti-skid tire tread. Then, the center coordinates of each area are calculated. The calculation method is as follows: (x, y) = (x t +0.5,y t +0.5), where x t is the horizontal coordinate of the centroid of the connected domain, y t is the ordinate of the centroid of the connected domain.

[0090] Since the center of the studded position of the anti-skid tire is a hole area, the coordinate data of this part of the area cannot be directly obtained when collecting data with a 3D camera. Therefore, in step S4, the pixel means Z1, Z2, Z3, Z4 and X1, X2 are calculated for the four areas selected around the same position in the height map and the X value map of each connected domain, so as to indirectly calculate the three-dimensional coordinates of the center point of the studded position area. The calculation method is as follows: Y = y × t,

[0091] According to the present invention, an automatic detection method for the studded position of an anti-skid tire tread is provided, by obtaining a brightness map and point cloud data of the anti-skid tire tread; processing the point cloud data to generate a height map and an X-value map; performing threshold segmentation on the brightness map to generate a binary image, then performing image sharpening processing on the height map, and then performing threshold segmentation on the processed result to generate a binary image. Then, each pixel point of the two binary images is compared to generate an image containing a minimum value, and then the pixels of the minimum value image are reversed; morphological processing is performed on the image; a connected domain analysis is performed on the processed image, and the connected domain is screened by using the feature difference between the studded position and other areas, and the result is displayed on the image; the coordinates of the center point of the screened connected domain in the image are calculated, and then the three-dimensional coordinates of the center of the studded position are calculated, and the coordinate data is saved in a required format file. The present invention can quickly and accurately detect and obtain the studded position of the anti-skid tire tread, which is convenient for the subsequent anti-skid tire manufacturing or modification process.

[0092] In order to realize the automatic detection method of the anti-skid tire tread stud position of the above-mentioned embodiment, the present invention also proposes an automatic detection system for the anti-skid tire tread stud position, and the detection system includes an acquisition module, a point cloud processing module, an image processing module, a display module and a storage module. The detection system corresponds to the detection method of the present invention and will not be described in detail here.

[0093] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A method for automatically detecting the position of studs on the tread of an anti-skid tire, characterized in that: The steps include: S1: Obtain the brightness image and point cloud data of the anti-skid tire tread, and perform threshold segmentation on the brightness image to generate a binary image; S2: Process the point cloud data to generate a height map and an X-value map, perform image sharpening on the height map, and then perform threshold segmentation to generate a binary image; S3: Compare the pixel values ​​of each pixel of the two binary images S1 and S2, retain the pixel with the smaller pixel value, and finally generate an image with the minimum pixel value, and then invert the pixels of the image with the minimum pixel value; S4: Perform morphological processing on the minimum pixel value image after inversion in S3; S5: Performing connected domain analysis on the image after morphological processing, screening connected domains by using feature differences between the pinning position and other regions in the connected domain that are not pinning regions, and displaying the screening results; S6: Calculate the coordinates of the center point of the connected domain after screening in the image, then calculate the three-dimensional coordinates of the center of the nail position, and save the position data in a required format file.

2. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 1, characterized in that: The point cloud data is processed to generate a height map and an X-value map, specifically: Extract the X, Y, and Z coordinates of each point in the point cloud data and define the grid resolution, whose size is determined by the brightness map resolution; The space is divided into grids, and the Z coordinate value and X coordinate value of each grid point in the point cloud are counted. The corresponding Z coordinate value and X coordinate value average of all points in each grid are taken, and the corresponding average values ​​of all grids are mapped to the two-dimensional image to generate the height map and X value map accordingly.

3. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 2, characterized in that: For a mesh with no points, set the value to -327.

4. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 1, characterized in that: The height map image is sharpened by defining a convolution kernel, and the parameters of the convolution kernel are set using the binary image of the generated brightness map.

5. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 4, characterized in that: The parameter value in the convolution kernel is determined by the corresponding position pixel in the binary image of the brightness map at the sliding position in the height map: When the pixel at this point in the binary image of the brightness map is 0, the value of the corresponding point of the convolution kernel is 0; When the pixel of this point in the binary image of the brightness map is 255, the value of this point in the convolution kernel is -1; If the midpoint of the convolution kernel exceeds the image range, the value of the out-of-range point is set to 0; The value of the center point of the convolution kernel is the number of points with a value of -1 in the convolution kernel.

6. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 5, characterized in that: Use the defined convolution kernel to traverse each pixel of the height map. If the pixel value of the corresponding point in the binary image of the brightness map is 0, the value of the point in the height map is replaced by 0; If the pixel in the middle of the binary image of the brightness image is 255, the value of the point in the height image is replaced by the calculated value. The calculation method is: the corresponding pixel point and the neighboring pixel are multiplied and accumulated with the convolution kernel and divided by the value of the center point of the convolution kernel plus 1; The pixel calculation formula for sharpening the height map is: in: The height map is H (x，y) , the binary image of the brightness image is I (x，y) , the convolution kernel is F (i，j) , whose size is (2m+1)×(2n+1), F (0，0) is the value of the center point of the convolution kernel corresponding to the calculated pixel.

7. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 1, characterized in that: The morphological processing is: First, the minimum image is eroded using a 3×3 rectangular structure element, then the minimum image is expanded using a 5×5 rectangular structure element, and finally the minimum image is eroded again using a 3×3 rectangular structure element.

8. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 1, characterized in that: The connected domain analysis is specifically as follows: S51) according to the centroid coordinates of each connected domain, filter and remove the connected domains whose pixel distance between the centroid and the image boundary is less than the length D; S52) calculating the area Area of ​​each connected domain, that is, the number of pixels contained in the connected domain, and filtering out the connected domains whose area is smaller than the minimum area threshold and whose area is larger than the maximum area threshold; S53) calculating the structural feature SF=A×B-1 of the connected domain, and filtering out the connected domains whose feature values ​​exceed 1; Where: A=R a / R b ; B = (π × R a ×R b ) / Area; A is the ratio of the major axis to the minor axis of each connected domain; B is the ratio of the product of the major axis, the minor axis and the pi of each connected domain to the area; S54) Calculate the inclination angles of each connected domain in the X and Y directions, select four regions with a length of w pixels and a width of h pixels in the left and right directions and the up and down directions of d pixels away from the centroid of each connected domain, calculate the mean values ​​of the regional pixels in the height map and the X value map, which are Z1, Z2, Z3, Z4 and X1, X2 respectively, and the calculation formula of the inclination angle of the connected domain is: Angel x =atan2(Z2-Z1,X2-X1),Angel y = atan2(Z4-Z3,2d×t); Among them, atan2 is the inverse tangent value, t is the resolution of the image in the Y direction, and the images with an inclination angle greater than the threshold θ in the X or Y direction are removed. x Or θ y connected domain.

9. The method for automatically detecting the position of studs on the tread of an anti-skid tire according to claim 8, characterized in that: The calculation process of the three-dimensional coordinates of the center of the nail position is: The remaining area after screening is the studded area in the anti-skid tire. The center point coordinates of the studded area in the anti-skid tire are obtained by the centroid coordinates of the screened connected domain. The calculation formula is: (x, y) = (x t +0.5,y t +0.5), where x t is the horizontal coordinate of the centroid of the connected domain, y t is the ordinate of the centroid of the connected domain; The three-dimensional coordinates of the center point are calculated according to the coordinates of the center point of the pinning position area in the minimum pixel value image, and the pixel average calculated for the selected area around each pinning position in the height map and the X value map. The calculation formula is: Y = y × t, 10. An automatic detection system for the position of studs on the tread of an anti-skid tire, characterized by: include A collection module, the collection module is used to collect brightness images and point cloud data of the anti-skid tire tread; A point cloud processing module, which is used to process point cloud data and generate a height map and an X-value map; An image processing module, wherein the image processing module is used to segment the brightness image and the height image to generate binary images, and then compare each pixel point of the binary image of the brightness image and the binary image of the height image to generate an image containing a minimum value, and then invert the pixels of the minimum pixel value image and perform morphological processing and connected domain analysis, and use the feature differences of other areas in the connected domain that are not the studded area to screen the connected domain, calculate the coordinates of the center point of the screened connected domain in the image, and calculate the three-dimensional coordinates of the center of the studded position; A display module, which is used to visualize the processing results and show the detection results of the nailing position to the user; The saving module is used to save the nailing position coordinate data obtained by detection in a required format and record the operation log of the system.

Images