Automobile tire pattern detection device based on machine vision and detection method thereof

Abstract

The invention relates to an automobile tire pattern detection device based on machine vision and a detection method thereof, and relates to the field of automobile part production and detection, and the automobile tire pattern detection device comprises a feeding assembly, a transportation assembly and a detection assembly. The feeding assembly comprises a base. And the feeding mechanical unit is movably connected with the base. The conveying assembly comprises a supporting unit, and a preset operation distance is reserved between the supporting unit and the base. The conveying rollers are connected with the supporting unit, and the number of the conveying rollers is multiple. The detection assembly comprises a positioning unit connected with the supporting unit; and the auxiliary limiting units are connected with the positioning unit, have preset operation sizes and are provided with a plurality of auxiliary limiting units. A corresponding visual detection part is arranged on the surface of the working face of the positioning unit, and corresponding imaging detection work can be carried out. Through mutual cooperation of the feeding assembly, the conveying assembly and the detection assembly, feeding treatment can be continuously carried out, defective products can be quickly sorted and discharged, and the overall efficiency of tire detection is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of automobile parts production detection, and in particular to an automobile tire pattern detection device and a detection method thereof based on machine vision. Background Art

[0002] Tires are circular elastic rubber products that roll on the ground and are installed on various vehicles or machines. They are usually installed on metal rims, can support the vehicle body, buffer external impacts, achieve contact with the road surface and ensure the driving performance of the vehicle. Tires are often used under complex and harsh conditions. They are subjected to various deformations, loads, forces, and high and low temperature effects when driving, so they must have high load-bearing performance, traction performance, and buffering performance. At the same time, they are also required to have high wear resistance and flex resistance, as well as low rolling resistance and heat generation.

[0003] Tires are an important part of the inspection process, whether in the production process or in the later maintenance process. The pattern on the tire surface, its standard level, wear degree, etc. are all closely related to the safety hazards of the car.

[0004] Traditional automobile tire pattern detection devices have a low loading speed and low detection efficiency. The selection of defective products requires manual assistance. At the same time, qualified products cannot be sorted from defective products in a timely manner, resulting in overall low efficiency. Summary of the invention

[0005] Purpose of the invention: To provide a vehicle tire pattern detection device based on machine vision, and further provide a detection method based on the above-mentioned vehicle tire pattern detection device based on machine vision, so as to solve the above-mentioned problems existing in the prior art.

[0006] Technical solution: A vehicle tire pattern detection device based on machine vision, including three components: a feeding component, a transportation component and a detection component.

[0007] The feeding assembly includes a base with a predetermined working size and supporting force; a feeding mechanical unit movably connected to the base and having a predetermined working length;

[0008] The transport assembly includes a support unit having a predetermined working length and supporting force and a predetermined working distance between the support unit and the base; a transport roller connected to the support unit and having a predetermined working length and supporting force, and the transport rollers are multiple;

[0009] The detection assembly includes a positioning unit connected to the support unit and having a predetermined operating size and a supporting force; a plurality of auxiliary limiting units connected to the positioning unit and having a predetermined operating size;

[0010] The working surface of the positioning unit is provided with a corresponding visual detection part, which can perform corresponding imaging detection operations.

[0011] In a further embodiment, the feeding mechanical unit includes four components: a rotating table, an adjusting arm, a connecting arm and a grabbing arm. The rotating table is rotatably connected to the base, has a predetermined operating size and a carrying space, and can perform a rotation adjustment operation within a predetermined range. The adjusting arm is rotatably connected to the rotating table, has a predetermined length and a supporting force, and can perform a rotation adjustment operation within a predetermined range. The connecting arm is rotatably connected to the adjusting arm, has a predetermined operating length and a supporting force, and can perform a rotation adjustment operation within a predetermined range. The grabbing arm is rotatably connected to the connecting arm, has a predetermined operating length and a supporting force, and can perform corresponding grabbing and transportation operations.

[0012] In a further embodiment, the grabbing arm includes a connecting plate, a limiting connecting plate and a grabbing plate; the connecting plate is rotatably connected to the connecting arm, has a predetermined working size, and can perform corresponding connection operations; the limiting connecting plate is connected to the connecting plate, and based on the predetermined working length, can perform corresponding limiting and bearing operations; the grabbing plate is slidably connected to the working surface of the limiting connecting plate, has a predetermined bending curvature, and is provided with corresponding auxiliary rollers; there are multiple grabbing plates, and they are relatively arranged on the working surface of the limiting connecting plate.

[0013] In a further embodiment, the support unit includes two components: a support column and a support frame. The support column has a predetermined operating size and supporting force, and there are multiple support columns, which can perform corresponding limiting and supporting operations; a predetermined operating distance is left between the multiple support columns. The support frame is connected to the support column, has a predetermined operating size and supporting force, is used for mutual communication between the multiple support columns, and can perform corresponding connection and limiting load-bearing operations.

[0014] In a further embodiment, the plurality of transport rollers are rotatably connected to the support frame, and a predetermined operating distance is left between the plurality of transport rollers, so that a rotation adjustment operation within a predetermined range can be performed. The plurality of support frames are the same, and the operating directions of the plurality of support frames are different, forming transport lines with different operating directions together with the support column and the transport rollers.

[0015] In a further embodiment, the positioning unit includes two components: a positioning column and a positioning frame. The positioning column is placed on the working surface of the support frame, has a predetermined working height and supporting force, and is multiple and symmetrically distributed on the working surface of the support frame. The positioning frame is connected to the positioning column, has a predetermined working size and bearing space, and is used for mutual communication between the multiple positioning columns.

[0016] In a further embodiment, the auxiliary limiting unit includes three components: a rotating connecting plate, an auxiliary rotating column and a mobile motor. The rotating connecting plate is placed on the working surface of the positioning frame, has a predetermined working size and supporting force, and one end is rotatably connected to the positioning frame, and can perform a predetermined range of rotation adjustment operations. The auxiliary rotating column passes through the rotating connecting plate and is movably connected to the rotating connecting plate, has a predetermined working length, and can perform a predetermined range of rotation adjustment operations. The rotating motor is placed on the working surface of the rotating connecting plate and is connected to the auxiliary rotating column through a transmission track, and can drive the auxiliary rotating column to perform corresponding rotation adjustment operations. The rotating connecting plate, the auxiliary connecting plate and the rotating motor are all multiple.

[0017] In a further embodiment, the visual inspection unit includes a detection positioning frame and an imaging detection camera; the detection positioning frame is placed on the working surface of the positioning frame, has a predetermined elasticity and supporting force, and can perform corresponding limiting and bearing operations; the imaging detection camera is connected to the detection positioning frame, and can perform corresponding imaging recording and analysis detection operations.

[0018] A detection method of a vehicle tire pattern detection device based on machine vision comprises the following steps:

[0019] S1. The automobile tire to be tested is grabbed by the feeding mechanical unit, and is prevented from being placed on the working surface of the transport roller, and is transported toward the direction close to the testing component via the transport roller;

[0020] S2. After the tire is transported to the predetermined working position, the plurality of auxiliary limiting units are rotated and adjusted in the direction close to the tire until the plurality of auxiliary limiting units are in contact with the tire to be inspected, and the auxiliary limiting units stop rotating;

[0021] S3, the detection positioning frame performs an extension operation, driving the imaging detection camera to adjust the corresponding working position, until the imaging detection camera moves to the predetermined working position and the detection positioning frame stops extending the operation;

[0022] S4, multiple auxiliary limit units are adjusted to rotate in the same direction, so that the tire rotates, and the imaging detection camera performs multi-directional imaging and recording with the tire during the rotation process, and detects whether the tire pattern is qualified through imaging analysis;

[0023] S5. The qualified products are further transported to the discharge port by the transport rollers; the defective products are transported to the defective product recovery port by the transport rollers with the assistance of the auxiliary limit unit.

[0024] In a further embodiment, the imaging detection camera needs to first input the standardized tire pattern information when performing imaging analysis, set it as the original comparison data, and collect image sets of different positions of the tire during the rotation process, the collected image sets are grayscale images, and then analyze the collected grayscale image sets. The specific analysis steps are as follows:

[0025] S401, performing denoising on the collected grayscale image, and converting the grayscale image into a binary image to distinguish between patterns and background;

[0026] Use filters to remove noise from an image:

[0027] I filtered =G(I,σ)

[0028] Where I is the original image, б is the standard deviation of the Gaussian filter;

[0029] The specific formula for converting a grayscale image into a binary image is as follows:

[0030]

[0031] Where T is the threshold;

[0032] S402, using the Canny edge detection algorithm to extract the edge of the pattern, the specific formula is as follows:

[0033] E=Canny(I gray ,σ,T1,T2)

[0034] Among them, б is the standard deviation of Gaussian filtering, T1, T2 are high and low thresholds;

[0035] Extract the outline of the pattern from the binary image. The specific formula is as follows:

[0036] C = FindContours(I binary )

[0037] Where C is the extracted contour set;

[0038] Harris corner points are used to detect the key feature points of the pattern. The specific formula is as follows:

[0039] P = Harris (I gray ,k,σ)

[0040] Among them, k is the parameter of Harris corner detection, and б is the standard deviation of Gaussian filtering;

[0041] S403, process the image to measure the depth, width and angle of the pattern, the specific formula is as follows:

[0042] D=Depth(Iscan )

[0043] Among them, I scan is the depth image captured by the imaging detection camera;

[0044] W=Width(C)

[0045] Among them, C is the outline of the pattern;

[0046] θ=Angle(C)

[0047] Among them, C is the outline of the pattern;

[0048] S404, identifying defects in the pattern and classifying the defects. The specific formula is as follows:

[0049] F=Defect(I binary )

[0050] Where F is the detected defect area;

[0051] L = Classify(F, Model)

[0052] Among them, L is the classification label of the defect, and Model is the trained classification model;

[0053] S405. Determine whether the tire is qualified based on the depth, width and angle parameters of the pattern. The specific formula is as follows:

[0054]

[0055] Among them, D min , W win ,θ min ,θ max It is the qualified standard of pattern;

[0056] S405: If the output value is 1, it means the tire meets the qualification standard; if the output value is 0, it means the tire does not meet the qualification standard.

[0057] Beneficial effects: The present invention relates to a vehicle tire pattern detection device based on machine vision and a detection method thereof, and relates to the field of automobile parts production and detection, including a feeding component, a transportation component and a detection component. The feeding component includes a base with a predetermined operating size and supporting force; a feeding mechanical unit movably connected to the base and having a predetermined operating length. The transportation component includes a support unit with a predetermined operating length and supporting force, and a predetermined operating distance between the base and the transportation roller connected to the support unit, and the transportation roller is multiple. The detection component includes a positioning unit connected to the support unit, with a predetermined operating size and supporting force; and an auxiliary limiting unit connected to the positioning unit, with a predetermined operating size and multiple. The working surface of the positioning unit is provided with a corresponding visual detection part, which can perform corresponding imaging detection operations. The present invention can continuously perform feeding processing through the mutual cooperation between the feeding component, the transportation component and the detection component, and can quickly sort and discharge defective products, effectively improving the overall efficiency of tire detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a top view of the present invention as a whole.

[0059] Figure 2 It is a schematic diagram of the feeding assembly described in the present invention.

[0060] Figure 3 Schematic diagram of some components of the present invention.

[0061] Figure 4 It is a side view of some components of the present invention.

[0062] The reference numerals in the figure are: feeding component 1, base 101, rotating table 102, adjusting arm 103, connecting arm 104, connecting plate 105, limiting connecting plate 106, grabbing plate 107, auxiliary roller 108, transport component 2, support column 201, support frame 202, transport roller 203, detection component 3, positioning column 301, positioning frame 302, rotating connecting plate 303, auxiliary rotating column 304, rotating motor 305, transmission crawler 306, detection positioning frame 307, imaging detection camera 308, tire 4. DETAILED DESCRIPTION

[0063] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0064] The applicant believes that the traditional automobile tire pattern detection device has a low loading speed and low detection efficiency. The selection of defective products requires manual assistance. At the same time, it is impossible to sort qualified products from defective products in a timely manner, resulting in overall low efficiency.

[0065] To this end, the applicant has designed a vehicle tire pattern detection device and method based on machine vision. Through the cooperation between the feeding component, transportation component and detection component, continuous feeding processing can be carried out, and defective products can be quickly sorted and unloaded, effectively improving the overall efficiency of tire detection.

[0066] The present invention relates to a machine vision-based automobile tire pattern detection device, which mainly includes three components: a feeding component 1, a transport component 2, and a detection component 3. The feeding component 1 includes a base 101 with a predetermined operating size and supporting force; a feeding mechanical unit movably connected to the base 101 and having a predetermined operating length. The transport component 2 includes a support unit with a predetermined operating length and supporting force, and a predetermined operating distance between the base 101 and the transport roller 203 connected to the support unit, and the transport roller 203 is multiple. The detection component 3 includes a positioning unit connected to the support unit, with a predetermined operating size and supporting force; and an auxiliary limiting unit connected to the positioning unit, with a predetermined operating size and multiple. The working surface of the positioning unit is provided with a corresponding visual detection part, which can perform corresponding imaging detection operations. In a further preferred embodiment, the support unit is multiple, and is interconnected with the multiple transport rollers 203, and the working directions of the multiple support units are different.

[0067] The feeding mechanical unit includes four components: a rotating table 102, an adjusting arm 103, a connecting arm 104 and a grabbing arm. The rotating table 102 is rotatably connected to the base 101, has a predetermined operating size and carrying space, and can perform a predetermined range of rotation adjustment operations. The adjusting arm 103 is rotatably connected to the rotating table 102, has a predetermined length and supporting force, and can perform a predetermined range of rotation adjustment operations. The connecting arm 104 is rotatably connected to the adjusting arm 103, has a predetermined operating length and supporting force, and can perform a predetermined range of rotation adjustment operations. The grabbing arm is rotatably connected to the connecting arm 104, has a predetermined operating length and supporting force, and can perform corresponding grabbing and transportation operations. In the later operation process, the rotating table 102 can be used to drive the adjusting arm 103 to adjust the corresponding operating direction as a whole, and the rotating adjustment of the adjusting arm 103 can drive the connecting arm 104 to adjust the corresponding operating height as a whole, so that the operating position of the grabbing arm can be accurately regulated to complete the corresponding tire 4 feeding operation. Repeat the above operation to achieve continuous tire 4 loading operation.

[0068] The grabbing arm includes a connecting plate 105, a limiting connecting plate 106 and a grab plate 107; the connecting plate 105 is rotatably connected to the connecting arm 104, has a predetermined working size, and can perform corresponding connection operations; the limiting connecting plate 106 is connected to the connecting plate 105, and can perform corresponding limiting and bearing operations based on a predetermined working length; the grab plate 107 is slidably connected to the working surface of the limiting connecting plate 106, has a predetermined bending arc, and is provided with a corresponding auxiliary roller 108; the grab plate 107 is multiple and is relatively arranged on the working surface of the limiting connecting plate 106. In a further preferred embodiment, the working surface of the auxiliary roller 108 is uneven, which can increase the friction between it and the tire 4, and avoid the occurrence of the tire 4 falling during the process of grabbing and feeding the tire 4.

[0069] The support unit includes two components: a support column 201 and a support frame 202. The support column 201 has a predetermined operating size and supporting force, and there are multiple support columns 201, which can perform corresponding limiting and supporting operations; a predetermined operating distance is left between the multiple support columns 201. The support frame 202 is connected to the support column 201, has a predetermined operating size and supporting force, is used for mutual communication between the multiple support columns 201, and can perform corresponding connection and limiting load-bearing operations.

[0070] The plurality of transport rollers 203 are rotatably connected to the support frame 202, and a predetermined operating distance is left between the plurality of transport rollers 203, so that a rotation adjustment operation within a predetermined range can be performed. The plurality of support frames 202 are the same, and the operating directions of the plurality of support frames 202 are different, forming transport lines with different operating directions with the support column 201 and the transport rollers 203. In a further preferred embodiment, the transport line formed by the plurality of support columns 201, the support frame 202 and the transport rollers 203 is T-shaped.

[0071] The positioning unit includes two components: a positioning column 301 and a positioning frame 302. The positioning column 301 is placed on the working surface of the support frame 202, has a predetermined working height and supporting force, and is multiple and symmetrically distributed on the working surface of the support frame 202. The positioning frame 302 is connected to the positioning column 301, has a predetermined working size and bearing space, and is used for mutual communication between the multiple positioning columns 301.

[0072] The auxiliary limiting unit includes three components: a rotating connecting plate 303, an auxiliary rotating column 304 and a moving motor. The rotating connecting plate 303 is placed on the working surface of the positioning frame 302, has a predetermined working size and supporting force, and one end is rotatably connected to the positioning frame 302, and can perform a predetermined range of rotation adjustment operations. The auxiliary rotating column 304 passes through the rotating connecting plate 303 and is movably connected to the rotating connecting plate 303, has a predetermined working length, and can perform a predetermined range of rotation adjustment operations. The rotating motor 305 is placed on the working surface of the rotating connecting plate 303, and is connected to the auxiliary rotating column 304 through a transmission track 306, and can drive the auxiliary rotating column 304 to perform corresponding rotation adjustment operations. The rotating connecting plate 303, the auxiliary connecting plate and the rotating motor 305 are all multiple. In the later operation process, a defective product transportation line is derived and set up on one side of the working position corresponding to the multiple auxiliary rotating columns 304. When defective products are detected in the later stage, the rotating connecting plate 303 away from the side of the defective product transportation line is synchronously rotated and adjusted toward the side close to the defective product transportation line, thereby driving the corresponding auxiliary rotating column 304 to move toward the end close to the defective product transportation line, and then the corresponding defective tire 4 can be pushed to the working surface of the defective product transportation line, and the defective product transportation line will perform corresponding transportation and unloading processing.

[0073] The visual inspection unit includes a detection positioning frame 307 and an imaging detection camera 308; the detection positioning frame 307 is placed on the working surface of the positioning frame 302, has a predetermined elasticity and supporting force, and can perform corresponding limiting and bearing operations; the imaging detection camera 308 is connected to the detection positioning frame 307, and can perform corresponding imaging recording and analysis detection operations.

[0074] On the basis of the above-mentioned depalletizing system, the present invention proposes a detection method of a vehicle tire pattern detection device based on machine vision, and the specific steps are as follows:

[0075] First, the automobile tire 4 to be inspected is grabbed by the loading mechanical unit, and is prevented from being placed on the working surface of the transport roller 203, and is transported toward the direction close to the inspection component 3 via the transport roller 203;

[0076] Next, when the tire 4 is transported to the predetermined working position, the plurality of auxiliary limiting units are rotated and adjusted in the direction close to the tire 4 until the plurality of auxiliary limiting units are in contact with the tire 4 to be inspected, and the auxiliary limiting units stop rotating;

[0077] Subsequently, the detection positioning frame 307 performs an extension operation, driving the imaging detection camera 308 to adjust the corresponding working position, until the imaging detection camera 308 moves to the predetermined working position and the detection positioning frame 307 stops the extension operation;

[0078] Then, the multiple auxiliary limit units are adjusted to rotate in the same direction, so that the tire 4 rotates, and the imaging detection camera 308 performs multi-directional imaging and recording with the tire 4 during the rotation process, and detects whether the pattern of the tire 4 is qualified through imaging analysis;

[0079] The qualified products are further transported to the discharge port by the transport roller 203 ; the defective products are transported to the defective product recovery port by the transport roller 203 with the assistance of the auxiliary limiting unit.

[0080] When performing imaging analysis, the imaging detection camera 308 needs to first input the standardized tire 4 pattern information, set it as the original comparison data, and collect image sets of the tire 4 at different positions during the rotation process. The collected image sets are grayscale images, and then analyze the collected grayscale image sets. The specific analysis steps are as follows:

[0081] S401, performing denoising on the collected grayscale image, and converting the grayscale image into a binary image to distinguish between patterns and background;

[0082] Use filters to remove noise from an image:

[0083] I filtered =G(I,σ)

[0084] Where I is the original image, б is the standard deviation of the Gaussian filter;

[0085] The specific formula for converting a grayscale image into a binary image is as follows:

[0086]

[0087] Where T is the threshold;

[0088] S402, using the Canny edge detection algorithm to extract the edge of the pattern, the specific formula is as follows:

[0089] E=Canny(I gray ,σ,T1,T2)

[0090] Among them, б is the standard deviation of Gaussian filtering, T1, T2 are high and low thresholds;

[0091] Extract the outline of the pattern from the binary image. The specific formula is as follows:

[0092] C = FindContours(I binary )

[0093] Where C is the extracted contour set;

[0094] Harris corner points are used to detect the key feature points of the pattern. The specific formula is as follows:

[0095] P = Harris (I gray ,k,σ)

[0096] Among them, k is the parameter of Harris corner detection, and б is the standard deviation of Gaussian filtering;

[0097] S403, process the image to measure the depth, width and angle of the pattern, the specific formula is as follows:

[0098] D=Depth(I scan )

[0099] Among them, I scan is the depth image captured by the imaging detection camera;

[0100] W=Width(C)

[0101] Among them, C is the outline of the pattern;

[0102] θ=Angle(C)

[0103] Among them, C is the outline of the pattern;

[0104] S404, identifying defects in the pattern and classifying the defects. The specific formula is as follows:

[0105] F=Defect(I binary )

[0106] Where F is the detected defect area;

[0107] L = Classify(F, Model)

[0108] Among them, L is the classification label of the defect, and Model is the trained classification model;

[0109] S405. Determine whether the tire is qualified based on the depth, width and angle parameters of the pattern. The specific formula is as follows:

[0110]

[0111] Among them, D min , W win ,θ min ,θ max It is the qualified standard of pattern;

[0112] S405: If the output value is 1, it means the tire meets the qualification standard; if the output value is 0, it means the tire does not meet the qualification standard.

[0113] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A vehicle tire pattern detection device based on machine vision, characterized in that include: A loading assembly, including a base having a predetermined working size and supporting force; A feeding mechanical unit movably connected to the base and having a predetermined operating length; A transport assembly, comprising a support unit having a predetermined operating length and a supporting force and a predetermined operating distance between the support unit and the base; A transport roller connected to the support unit and having a predetermined working length and supporting force, wherein the transport roller is multiple; The detection assembly includes a positioning unit connected to the support unit and having a predetermined operating size and a supporting force; and a plurality of auxiliary limiting units connected to the positioning unit and having a predetermined operating size; The working surface of the positioning unit is provided with a corresponding visual detection part, which can perform corresponding imaging detection operations.

2. The device for detecting automobile tire pattern based on machine vision according to claim 1, characterized in that: The feeding mechanical unit comprises: A rotating platform is rotatably connected to the base, has a predetermined operating size and carrying space, and can perform a rotation adjustment operation within a predetermined range; An adjustment arm is rotatably connected to the rotating platform, has a predetermined length and a supporting force, and can perform a rotation adjustment operation within a predetermined range; A connecting arm is rotatably connected to the adjusting arm, has a predetermined operating length and supporting force, and can perform a rotation adjustment operation within a predetermined range; The grabbing arm is rotatably connected to the connecting arm and has a predetermined operating length and supporting force, and can perform corresponding grabbing and transporting operations.

3. The device for detecting automobile tire pattern based on machine vision according to claim 2, characterized in that: The grabbing arm includes a connecting plate, a limiting connecting plate and a grabbing plate; the connecting plate is rotatably connected to the connecting arm, has a predetermined working size, and can perform corresponding connection operations; the limiting connecting plate is connected to the connecting plate, and based on the predetermined working length, can perform corresponding limiting and bearing operations; the grabbing plate is slidably connected to the working surface of the limiting connecting plate, has a predetermined bending curvature, and is provided with corresponding auxiliary rollers; there are multiple grabbing plates, which are relatively arranged on the working surface of the limiting connecting plate.

4. The device for detecting automobile tire pattern based on machine vision according to claim 1, characterized in that: The support unit comprises: The supporting columns have a predetermined operating size and supporting force, and are multiple in number, and can perform corresponding limiting and supporting operations; a predetermined operating distance is left between the multiple supporting columns; The support frame is connected to the support column, has a predetermined operating size and supporting force, is used for interconnection between multiple support columns, and can perform corresponding connection and limited load-bearing operations.

5. The device for detecting automobile tire pattern based on machine vision according to claim 4, characterized in that: The plurality of transport rollers are rotatably connected to the support frame, and a predetermined operating distance is left between the plurality of transport rollers, so that a rotation adjustment operation within a predetermined range can be performed; There are multiple support frames, and the operating directions of the multiple support frames are different, and they form transport lines with different operating directions together with the support columns and transport rollers.

6. The device for detecting automobile tire pattern based on machine vision according to claim 4, characterized in that: The positioning unit comprises: A positioning column, which is placed on the working surface of the support frame, has a predetermined working height and supporting force, and is in multiple numbers and symmetrically distributed on the working surface of the support frame; The positioning frame is connected to the positioning posts, has a predetermined operating size and a bearing space, and is used for mutual communication between the plurality of positioning posts.

7. The device for detecting automobile tire pattern based on machine vision according to claim 6, characterized in that: The auxiliary limiting unit comprises: A rotating connecting plate is placed on the working surface of the positioning frame, has a predetermined working size and supporting force, and one end is rotatably connected to the positioning frame, and can perform a rotation adjustment operation within a predetermined range; An auxiliary rotating column passes through the rotating connecting plate and is movably connected to the rotating connecting plate, has a predetermined operating length, and can perform a rotating adjustment operation within a predetermined range; A rotating motor is placed on the working surface of the rotating connecting plate and is connected to the auxiliary rotating column through a transmission track, and can drive the auxiliary rotating column to perform corresponding rotation adjustment operations; The rotating connecting plate, the auxiliary connecting plate and the rotating motor are all multiple.

8. The device for detecting automobile tire pattern based on machine vision according to claim 6, characterized in that: The visual inspection unit includes a detection positioning frame and an imaging detection camera; the detection positioning frame is placed on the working surface of the positioning frame, has a predetermined elasticity and supporting force, and can perform corresponding limiting and bearing operations; The imaging detection camera is connected to the detection positioning frame and can perform corresponding imaging recording and analysis detection operations.

9. The detection method of the automobile tire pattern detection device based on machine vision according to any one of claims 1 to 8, characterized in that The following steps are involved: S1. The automobile tire to be tested is grabbed by the feeding mechanical unit, and is prevented from being placed on the working surface of the transport roller, and is transported toward the direction close to the testing component via the transport roller; S2. After the tire is transported to the predetermined working position, the plurality of auxiliary limiting units are rotated and adjusted in the direction close to the tire until the plurality of auxiliary limiting units are in contact with the tire to be inspected, and the auxiliary limiting units stop rotating; S3, the detection positioning frame performs an extension operation, driving the imaging detection camera to adjust the corresponding working position, until the imaging detection camera moves to the predetermined working position and the detection positioning frame stops extending the operation; S4, multiple auxiliary limit units are adjusted to rotate in the same direction, so that the tire rotates, and the imaging detection camera performs multi-directional imaging and recording with the tire during the rotation process, and detects whether the tire pattern is qualified through imaging analysis; S5. The qualified products are further transported to the discharge port by the transport rollers; the defective products are transported to the defective product recovery port by the transport rollers with the assistance of the auxiliary limit unit.

10. The detection method of the automobile tire pattern detection device based on machine vision according to claim 1, characterized in that: When performing imaging analysis, the imaging detection camera needs to first input the standardized tire pattern information, set it as the original comparison data, and collect image sets of different positions of the tire during the rotation process. The collected image sets are grayscale images, and then analyze the collected grayscale image sets. The specific analysis steps are as follows: S401, performing denoising on the collected grayscale image, and converting the grayscale image into a binary image to distinguish between patterns and background; Use filters to remove noise from an image: I filtered =G(I,σ) Where I is the original image, б is the standard deviation of the Gaussian filter; The specific formula for converting a grayscale image into a binary image is as follows: Where T is the threshold; S402, using the Canny edge detection algorithm to extract the edge of the pattern, the specific formula is as follows: E=Canny(I gray ,σ,T1,T2) Among them, б is the standard deviation of Gaussian filtering, T1, T2 are high and low thresholds; Extract the outline of the pattern from the binary image. The specific formula is as follows: C=FindContours(I binary ) Where C is the extracted contour set; Harris corner points are used to detect the key feature points of the pattern. The specific formula is as follows: P=Harris(I gray ,k,σ) Among them, k is the parameter of Harris corner detection, and б is the standard deviation of Gaussian filtering; S403, process the image to measure the depth, width and angle of the pattern, the specific formula is as follows: D=Depth(I scan ) Among them, I scan is the depth image captured by the imaging detection camera; W=Width(C) Among them, C is the outline of the pattern; θ=Angle(C) Among them, C is the outline of the pattern; S404, identifying defects in the pattern and classifying the defects. The specific formula is as follows: F=Defect(I binary ) Where F is the detected defect area; L = Classify(F, Model) Among them, L is the classification label of the defect, and Model is the trained classification model; S405. Determine whether the tire is qualified based on the depth, width and angle parameters of the pattern. The specific formula is as follows: Among them, D min , W win ,θ min ,θ max It is the qualified standard of pattern; S405: If the output value is 1, it means the tire meets the qualification standard; if the output value is 0, it means the tire does not meet the qualification standard.

Images