Image Recognition-Based Intelligent Intubation Control Method and System for Motor Vehicle Exhaust Gas Detection

Abstract

This invention relates to the field of exhaust pipe insertion control technology, specifically to an intelligent exhaust pipe insertion control method and system for motor vehicle exhaust gas detection based on image recognition. To address the technical problems of low intelligence and poor accuracy in existing motor vehicle exhaust pipe insertion detection technologies, this invention converts the two-dimensional coordinates of a vehicle rear image into the three-dimensional coordinates of a robotic arm, determines the position of the exhaust pipe in the robotic arm's workspace, and plans the movement path based on the joint angle information of the robotic arm. Then, according to the robotic arm's movement path, the exhaust gas collection pipe at the end of the robotic arm is positioned directly above the exhaust pipe. A segmented speed control method is employed during the insertion of the exhaust gas collection pipe, and the insertion pressure is acquired in real time. If the pressure exceeds a pressure threshold, the pipe is pulled out and re-inserted. This insertion control method exhibits strong adaptability and high intelligence, achieving a high degree of automation in the exhaust gas detection insertion process, thereby improving the accuracy and efficiency of exhaust gas detection.

Description

Technical Field

[0001] This invention relates to the field of exhaust pipe insertion control technology, specifically to an intelligent exhaust pipe detection control method and system based on image recognition for motor vehicles. Background Technology

[0002] During vehicle exhaust emission testing, it is usually necessary to insert the exhaust gas collection pipe into the vehicle's exhaust pipe. Currently, this operation is mainly carried out manually, which is not only labor-intensive, but also difficult to guarantee the insertion depth, resulting in significant errors and affecting the accuracy of the exhaust emission test results. In addition, operators frequently approach the exhaust outlet and inhale exhaust pollutants, which also has a significant adverse impact on their health. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent intubation control method and system for detecting motor vehicle exhaust emissions based on image recognition.

[0004] The technical solution of this invention is as follows:

[0005] A method for intelligent intubation control of motor vehicle exhaust emission detection based on image recognition includes the following operations:

[0006] S1. Acquire an image of the rear of the car and identify the exhaust pipe from the image of the rear of the car;

[0007] S2. After converting the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, path planning is performed based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm. The specific method for obtaining the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm is as follows: Based on the intrinsic and extrinsic parameters of the camera, a camera imaging model and a transformation relationship model between the camera coordinate system and the robot arm coordinate system are constructed. After substituting the two-dimensional coordinates of the exhaust pipe into the camera imaging model, the model is processed by the least squares method to obtain the world vertical coordinates of the exhaust pipe. The world vertical coordinates of the exhaust pipe and the two-dimensional coordinates of the exhaust pipe are substituted into the camera imaging model for solving to obtain the world horizontal coordinates and the world vertical coordinates of the exhaust pipe. The world horizontal coordinates, world vertical coordinates, and world vertical coordinates of the exhaust pipe form the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm.

[0008] S3. According to the robot arm's movement path, the exhaust gas collection pipe at the end of the robot arm is aligned with the exhaust pipe. Using a segmented speed control method, the exhaust gas collection pipe is inserted into the exhaust pipe to a preset depth threshold in a preset insertion direction. During the insertion of the exhaust gas collection pipe, the insertion pressure of the exhaust gas collection pipe is acquired in real time. If the insertion pressure of the exhaust gas collection pipe is greater than the pressure threshold, the exhaust gas collection pipe is pulled out to the pull-out depth threshold in the opposite direction to the insertion direction, and then the exhaust gas collection pipe insertion action is performed.

[0009] The specific steps in S2 to obtain the robot arm's movement path are as follows: The initial angles of each joint of the robot arm are used as the root nodes of a random tree to obtain an initial random tree; the initial random tree undergoes node expansion iterations until the distance between a newly generated node and the target angle point is less than a distance threshold, at which point the iteration ends; the new node whose distance to the target angle point is less than the distance threshold is used as the target node; during the backtracking process from the target node to the root node, the edges between nodes in the random tree are traversed to obtain several movement paths; based on the path cost function, these movement paths are filtered, and the movement path with the lowest path cost is used as the robot arm's movement path; the target angle point is the point at the end of the robot arm corresponding to the standard exhaust pipe position.

[0010] The path cost function is obtained through the following formula:

[0011] ,

[0012] Q Value for the path d(p i ,p i+1 ) For nodes p i With nodes p i+1 The distance between them i i+1 For nodes p i+1 Angle, i i For nodes p i Angle, q e Based on the energy consumption of the robot arm during its movement along the motion path I This represents the total number of nodes in the motion path.

[0013] During the node expansion iteration process, the specific operations of the current node expansion process are as follows: In the random tree generated in the previous round, a leaf node is randomly generated within the range of the minimum and maximum movement angles of each root node, resulting in several current transition leaf nodes; the node in the random tree generated in the previous round that is closest to the corresponding current transition leaf node is taken as the current nearest leaf node of each root node; the position point of the current nearest leaf node after moving along the direction of the corresponding current transition leaf node by a step size threshold is taken as the current new node of the corresponding root node; the current new node is connected to the corresponding current nearest leaf node to realize the update iteration of the random tree, obtaining the random tree generated in the current round, which is used to execute the node expansion process in the next round.

[0014] In S1, the operation of identifying the exhaust pipe from the rear image of a car is as follows: Edge points in the rear image of the car are marked to obtain a marked rear image; detection boxes are filled onto the marked rear image, and the detection boxes corresponding to the minimum distance between the center point of the detection box and the center point of the ground truth box within the same region are used as the classification boxes for the corresponding regions; based on the shape, pixel distribution, and area of ​​the exhaust pipe, all classification boxes are classified, and the classification boxes that are classified as exhaust pipes and whose corresponding classification probability value is greater than a probability threshold are used as exhaust pipe recognition boxes; the exhaust pipe recognition boxes are processed by Hough transform to obtain the center position of the exhaust pipe; the coordinates of the center position of the exhaust pipe in the rear image of the car are used as the two-dimensional coordinates of the exhaust pipe.

[0015] The method for obtaining edge location points is as follows: the rear image of the car is converted to grayscale to obtain a grayscale image of the car; the weighted average of the pixels of each location point and other location points within the first neighborhood in the grayscale image of the car is taken as the new pixel value of the corresponding location point to obtain a denoised image of the car; it is determined whether the pixel value of each location point in the denoised image of the car is greater than the edge pixel threshold; if the pixel value of the current location point is not less than the first edge pixel threshold, then the current location point is an edge location point; if the pixel value of the current location point is between the first edge pixel threshold and the second edge pixel threshold, and the current location point is adjacent to the determined edge location point, then the current location point is an edge location point; the first edge pixel threshold is greater than the second edge pixel threshold.

[0016] An image recognition-based intelligent intubation control system for motor vehicle exhaust emission detection, used to implement the aforementioned image recognition-based intelligent intubation control method for motor vehicle exhaust emission detection, includes:

[0017] The robotic arm has an exhaust gas collection pipe at its end, which is used to move the exhaust gas collection pipe according to the instructions transmitted from the control unit.

[0018] A camera, mounted on a robotic arm, is used to capture images of the rear of a car.

[0019] The image processing unit is used to process the rear-end image of the car captured by the camera and identify the exhaust pipe from the rear-end image of the car.

[0020] The computing unit is used to convert the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, and then perform path planning based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm.

[0021] The control unit is used to move the exhaust gas collection pipe at the end of the robot arm to face the exhaust pipe according to the robot arm movement path in the computing unit; insert the exhaust gas collection pipe into the exhaust pipe to a preset depth threshold using a segmented speed control method in a preset insertion direction; and when the exhaust gas collection pipe insertion pressure is greater than the pressure threshold during the insertion process, pull out the exhaust gas collection pipe to the pull-out depth threshold in the opposite direction to the insertion direction and then perform the exhaust gas collection pipe insertion action.

[0022] Pressure sensor used to obtain the insertion pressure of the exhaust gas collection pipe;

[0023] A depth sensor is used to determine the depth at which the exhaust gas collection pipe is inserted into the exhaust pipe;

[0024] The exhaust gas collection pipe fixing clamp is installed at the end of the robot's robotic arm and is used to fix the exhaust gas collection pipe.

[0025] The system includes two cameras: one mounted on the main support of the robot arm to capture images of the car, and the other mounted at the end of the robot arm to capture images of the rear of the car when the rear of the car is visible in the image.

[0026] The beneficial effects of this invention are as follows: This invention provides an intelligent tube insertion control method for motor vehicle exhaust emission detection based on image recognition. First, the position and shape of the exhaust pipe at the rear of the vehicle are determined by image recognition of the exhaust pipe. Then, the position of the exhaust pipe in the workspace of the robotic arm is determined by converting the two-dimensional coordinates of the rear image of the vehicle into the three-dimensional coordinates of the robotic arm. The movement path is planned by combining the joint angle information of the robotic arm, providing clear motion commands to the robotic arm. Finally, according to the movement path of the robotic arm, the exhaust gas collection tube at the end of the robotic arm is positioned directly above the exhaust pipe. A segmented speed control method is used during the insertion of the exhaust gas collection tube, which better controls the insertion accuracy. Simultaneously, the insertion pressure of the exhaust gas collection tube is acquired in real time, and if the pressure exceeds a pressure threshold, the tube is pulled out and re-inserted, effectively preventing damage to the exhaust gas collection tube or exhaust pipe due to excessive insertion force. This tube insertion control method has strong adaptability and high intelligence, achieving a high degree of automation in the exhaust emission detection tube insertion process, reducing the tediousness and uncertainty of manual operation, and improving the accuracy and efficiency of exhaust emission detection. Detailed Implementation

[0027] This embodiment provides an image recognition-based intelligent intubation control system for motor vehicle exhaust emission detection, used to implement an image recognition-based intelligent intubation control method for motor vehicle exhaust emission detection, including:

[0028] The robotic arm has an exhaust gas collection pipe at its end, which is used to move the exhaust gas collection pipe according to the instructions transmitted from the control unit.

[0029] The camera is mounted on the robotic arm to acquire images of the rear of a car. There are two cameras: one is mounted on the main post of the robotic arm to acquire images of the car, and the other is mounted at the end of the robotic arm to acquire images of the rear of the car when the image processing unit outputs an image of the car, thereby improving recognition accuracy.

[0030] The image processing unit is used to process the rear-end image of the car captured by the camera and identify the exhaust pipe from the rear-end image of the car.

[0031] The computing unit is used to convert the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, and then perform path planning based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm.

[0032] The control unit is used to move the exhaust gas collection pipe at the end of the robot arm to face the exhaust pipe according to the robot arm movement path in the computing unit; insert the exhaust gas collection pipe into the exhaust pipe to a preset depth threshold using a segmented speed control method in a preset insertion direction; and when the exhaust gas collection pipe insertion pressure is greater than the pressure threshold during the insertion process, pull out the exhaust gas collection pipe to the pull-out depth threshold in the opposite direction to the insertion direction, or pull out the exhaust gas collection pipe to the corresponding pull-out depth threshold, and then perform the exhaust gas collection pipe insertion action.

[0033] Pressure sensor used to obtain the insertion pressure of the exhaust gas collection pipe;

[0034] A depth sensor is used to determine the depth at which the exhaust gas collection pipe is inserted into the exhaust pipe;

[0035] Exhaust gas collection pipe fixing clamp, installed at the end of the robot arm, is used to fix the exhaust gas collection pipe;

[0036] The exhaust gas collection pipe retraction device is installed at the end of the robot arm and is connected to the exhaust gas collection pipe. It is used for the automatic retraction of the exhaust gas collection pipe during the movement of the robot arm.

[0037] This embodiment also provides an intelligent intubation control method for motor vehicle exhaust emission detection based on image recognition, the specific process of which is as follows.

[0038] S1. Acquire an image of the rear of the car and identify the exhaust pipe from the image.

[0039] By using image recognition of the exhaust pipe, the position and shape of the exhaust pipe at the rear of the car can be accurately determined, improving the level of automation and helping to improve the accuracy of subsequent pipe insertion.

[0040] The specific steps for identifying the exhaust pipe from a car rear image are as follows: First, mark the edge points in the car rear image to obtain a marked car rear image that is easy to identify accurately and quickly. Second, fill the marked car rear image with detection boxes, and use the detection box whose center point in the same area has the minimum distance from the center point of the ground truth box as the corresponding classification box. Third, classify all classification boxes based on the shape, pixel distribution, and area of ​​the exhaust pipe. Use the classification box that identifies the exhaust pipe and whose corresponding classification probability value is greater than a probability threshold as the exhaust pipe recognition box. Fourth, perform a Hough transform on the exhaust pipe recognition box to obtain the center position of the exhaust pipe. Finally, use the coordinates of the exhaust pipe center position in the car rear image as the two-dimensional coordinates of the exhaust pipe.

[0041] The method for obtaining edge location points is as follows: The rear image of the car is converted to grayscale to obtain a grayscale image of the car; the weighted average of the pixels at each location point and other locations within a first neighborhood in the grayscale image of the car is used as the new pixel value for the corresponding location point, resulting in a denoised image of the car; it is determined whether the pixel value of each location point in the denoised image of the car is greater than an edge pixel threshold; if the pixel value of the current location point is not less than the first edge pixel threshold, then the current location point is an edge location point; if the pixel value of the current location point is between the first edge pixel threshold and the second edge pixel threshold, and the current location point is adjacent to a determined edge location point, then the current location point is an edge location point. The first edge pixel threshold is greater than the second edge pixel threshold.

[0042] The method for obtaining edge location points can also be specifically as follows: Convert the car rear image to grayscale to obtain a car grayscale image; take the weighted average of the pixels at each location point and other locations within the first neighborhood in the car grayscale image as the new pixel value for the corresponding location point to obtain a denoised car image; obtain the pixel gradient magnitude and pixel gradient direction at each location point in the denoised car image; compare the pixel gradient magnitude of each location point with the gradient magnitudes of all adjacent pixels within the second neighborhood and along the corresponding pixel gradient direction; if the pixel gradient magnitude of the current location point is the maximum value, then the current location point is an edge location point.

[0043] S2. After converting the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, path planning is performed based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm.

[0044] By converting the two-dimensional coordinates of the car's rear image into the three-dimensional coordinates of the robotic arm, the position of the exhaust pipe in the robotic arm's workspace can be accurately determined. Combined with the joint angle information of the robotic arm, the movement path is planned, providing the robotic arm with clear motion commands. This achieves a high degree of automation in the exhaust gas detection and pipe insertion process, reduces the tediousness and uncertainty of manual operation, and improves the automation level of the entire detection process.

[0045] First, the two-dimensional coordinates of the car's rear image are converted into the three-dimensional coordinates of the robot's operation. This transforms the two-dimensional coordinates of the exhaust pipe into the three-dimensional coordinates of the robot arm, obtaining the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm, which facilitates subsequent robot arm movement path planning.

[0046] The method for obtaining the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm is as follows: Based on the intrinsic and extrinsic parameters of the camera, a camera imaging model and a transformation relationship model between the camera coordinate system and the robot arm coordinate system (world coordinate system) are constructed; after substituting the two-dimensional coordinates of the exhaust pipe into the camera imaging model, the world vertical coordinates of the exhaust pipe are obtained by processing them with the transformation relationship model using the least squares method; the world vertical coordinates of the exhaust pipe and the two-dimensional coordinates of the exhaust pipe are substituted into the camera imaging model for solving to obtain the world horizontal coordinates and the world vertical coordinates of the exhaust pipe; the world horizontal coordinates, the world vertical coordinates of the exhaust pipe, and the world vertical coordinates of the exhaust pipe form the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm during operation.

[0047] Then, based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm, the movement path of the robot arm is obtained.

[0048] The specific steps for obtaining the robot arm's movement path are as follows: The initial angles of each joint of the robot arm are used as the root nodes of a random tree to obtain an initial random tree; the initial random tree undergoes iterative expansion until the distance between a newly generated node and the target angle point is less than a distance threshold, at which point the iteration ends; the new node whose distance to the target angle point is less than the distance threshold is used as the target node; during the backtracking process from the target node to the root node, the edges between nodes in the random tree are traversed to construct a motion path from the initial state to the target state, resulting in several motion paths; based on the path cost function, these motion paths are filtered, and the motion path with the lowest path cost is used as the robot arm's movement path.

[0049] The target angle point mentioned above is the point at the end of the robot arm corresponding to the standard exhaust pipe position, that is, the position point where the exhaust gas collection pipe at the end of the robot arm is directly opposite the exhaust pipe. It is also the initial position point of the end of the robot arm when the insertion process is performed. The target angle point can be calculated based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm during operation.

[0050] In the above node expansion iteration process, the specific operation of the current node expansion process is as follows: In the random tree generated in the previous round, a leaf node is randomly generated within the range of the minimum and maximum movement angles of each root node, resulting in several current transition leaf nodes; each current transition leaf node corresponds to its respective root node; the node in the random tree generated in the previous round that is closest to the corresponding current transition leaf node is taken as the current nearest leaf node for each root node; the position point of the current nearest leaf node after moving along the direction of the corresponding current transition leaf node by a step size threshold is taken as the current new node of the corresponding root node; the current new node is connected to the corresponding current nearest leaf node (the current new node belonging to the same root node is connected to the current nearest leaf node), thereby updating and iterating the random tree to obtain the random tree generated in the current round, which is used to execute the node expansion process in the next round. The step size threshold is obtained based on the length of the robotic arm and the range of working angles.

[0051] Meanwhile, the path cost function is obtained through the following formula:

[0052] ,

[0053] Q Value for the path d(p i ,p i+1 ) For nodes p i With nodes p i+1 The distance between them i i+1 For nodes p i+1 Angle, i i For nodes p i Angle, q e Based on the energy consumption of the robot arm during its movement along the motion path I This represents the total number of nodes in the motion path.

[0054] The operation to obtain the robot arm's movement path can also be as follows: determine whether the position point of the exhaust pipe in the robot arm's three-dimensional coordinates is within the movement range of the robot arm's end effector; if it is, then the line connecting the position point of the exhaust pipe in the robot arm's three-dimensional coordinates with the position point of the robot arm's end effector corresponding to the standard exhaust pipe position is taken as the robot arm's movement path; if it is not, then the point projected from the initial position point of the robot arm's end effector onto the horizontal plane where the position point of the exhaust pipe in the robot arm's three-dimensional coordinates is located is taken as the midpoint, and the 90° broken line formed by the initial position point of the robot arm's end effector, the midpoint, and the position point of the robot arm's end effector corresponding to the standard exhaust pipe position is taken as the robot arm's movement path.

[0055] S3. According to the robot arm's movement path, the exhaust gas collection pipe at the end of the robot arm is aligned with the exhaust pipe. Using a segmented speed control method, the exhaust gas collection pipe is inserted into the exhaust pipe to a preset depth threshold in a preset insertion direction. During the insertion of the exhaust gas collection pipe, the insertion pressure of the exhaust gas collection pipe is acquired in real time. If the insertion pressure of the exhaust gas collection pipe is greater than the pressure threshold, the exhaust gas collection pipe is pulled out to the pull-out depth threshold in the opposite direction to the insertion direction, and then the exhaust gas collection pipe insertion action is performed.

[0056] Based on the robot arm's movement path, the exhaust gas collection pipe at the end of the moving robot arm is aligned with the exhaust pipe. During the insertion of the exhaust gas collection pipe, a segmented speed control method is adopted, which can better control the insertion accuracy. At the same time, the insertion pressure of the exhaust gas collection pipe is acquired in real time, and measures are taken to pull out and re-insert when the pressure exceeds the pressure threshold. This can effectively prevent the exhaust gas collection pipe or exhaust pipe from being damaged due to excessive insertion force, improve the adaptability of the insertion method, enhance the ability to cope with complex situations, and improve the intelligence of the insertion method.

[0057] First, based on the robot arm's movement path, the exhaust gas collection pipe at the end of the robot arm is aligned with the exhaust pipe, that is, from the end of the robot arm to the target angle point, the exhaust gas collection pipe is aligned with the exhaust pipe.

[0058] Then, using a segmented speed control method with a preset insertion direction, the exhaust gas collection tube is inserted into the exhaust pipe to a preset depth threshold (preferably 350mm). In this segmented speed control method, a decreasing speed principle is adopted: the speed approaching the exhaust pipe (e.g., 1-2 cm / s) is greater than the speed during insertion (e.g., 0.5-1 cm / s). The relatively faster speed during the approach phase allows the exhaust gas collection tube to quickly approach the exhaust pipe, reducing the overall insertion time. However, as it approaches the exhaust pipe, the speed is reduced to 0.5-1 cm / s for insertion. This slower speed allows the robotic arm to more precisely control the position of the exhaust gas collection tube, ensuring accurate alignment between the tube and the exhaust pipe opening. This helps avoid insertion position deviations caused by excessive speed, thereby improving insertion accuracy.

[0059] Meanwhile, during the insertion of the exhaust gas collection pipe, the insertion pressure of the exhaust gas collection pipe is acquired in real time. If the insertion pressure of the exhaust gas collection pipe is greater than the pressure threshold, it indicates that the exhaust gas collection pipe may be stuck. In this case, the exhaust gas collection pipe is pulled out to the extraction depth threshold in the opposite direction to the insertion direction, or pulled out to the extraction depth threshold. Then, the exhaust gas collection pipe insertion action is performed according to the corresponding insertion direction and speed to continue inserting the exhaust gas collection pipe into the exhaust pipe.

[0060] This embodiment provides an intelligent tube insertion control method for motor vehicle exhaust emission detection based on image recognition. First, the exhaust pipe's position and shape at the rear of the vehicle are determined through image recognition. Then, the two-dimensional coordinates of the vehicle's rear image are converted into three-dimensional coordinates for a robotic arm, determining the exhaust pipe's position within the robotic arm's workspace. A movement path is planned using the robotic arm's joint angle information, providing clear motion commands to the robotic arm. Finally, based on the robotic arm's movement path, the exhaust gas collection tube at the end of the robotic arm is positioned directly above the exhaust pipe. A segmented speed control method is employed during the insertion process to better control insertion accuracy. Simultaneously, the insertion pressure of the exhaust gas collection tube is acquired in real time, and if the pressure exceeds a threshold, the tube is pulled out and re-inserted, effectively preventing damage to the exhaust gas collection tube or exhaust pipe due to excessive insertion force. This tube insertion control method exhibits strong adaptability and high intelligence, achieving a high degree of automation in the exhaust emission detection tube insertion process. It reduces the tediousness and uncertainty of manual operation, improving the accuracy and efficiency of exhaust emission detection.

Claims

1. A method for intelligent intubation control of motor vehicle exhaust emission detection based on image recognition, characterized in that, This includes the following operations: S1. Acquire an image of the rear of the car and identify the exhaust pipe from the image of the rear of the car; S2. After converting the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, path planning is performed based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm. The method for obtaining the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm is as follows: Based on the intrinsic and extrinsic parameters of the camera, a camera imaging model and a transformation relationship model between the camera coordinate system and the robot arm coordinate system are constructed; after substituting the two-dimensional coordinates of the exhaust pipe into the camera imaging model, the model is processed by the least squares method to obtain the world vertical coordinates of the exhaust pipe; the world vertical coordinates of the exhaust pipe and the two-dimensional coordinates of the exhaust pipe are substituted into the camera imaging model for solving to obtain the world horizontal coordinates and the world vertical coordinates of the exhaust pipe; the world horizontal coordinates, world vertical coordinates, and world vertical coordinates of the exhaust pipe form the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm during operation. In path planning, the initial angles of each joint of the robot arm are used as the root node of the random tree to obtain the initial random tree; The initial random tree undergoes node expansion iterations until the distance between the newly generated node and the target angle point is less than a distance threshold, at which point the iteration process ends. New nodes whose distance to the target angle point is less than the distance threshold are designated as target nodes; During the backtracking process from the target node to the root node, the edges between nodes in the random tree are traversed to obtain several motion paths; the motion path with the minimum path cost is taken as the movement path of the robot arm. The target angle point is the point at the end of the robotic arm corresponding to the standard exhaust pipe position; In the current node expansion, a leaf node is randomly generated within the range of the minimum and maximum movement angles of each root node in the random tree generated in the previous round, resulting in several current transition leaf nodes; the node in the random tree generated in the previous round that is closest to the corresponding current transition leaf node is taken as the current nearest leaf node of each root node. The position of the current nearest leaf node after moving along the direction of the corresponding current transition leaf node by a step size threshold is taken as the current new node of the corresponding root node; the current new node is connected with the corresponding current nearest leaf node to obtain the random tree generated in the current round, which is used to perform the node expansion processing operation in the next round; S3. According to the movement path of the robot arm, the exhaust gas collection pipe at the end of the robot arm is aligned with the exhaust pipe. With a preset insertion direction, the exhaust gas collection pipe is inserted into the exhaust pipe to a preset depth threshold using a segmented speed control method, with the speed decreasing progressively. During the insertion of the exhaust gas collection pipe, the insertion pressure of the exhaust gas collection pipe is acquired in real time. If the insertion pressure of the exhaust gas collection pipe is greater than the pressure threshold, the exhaust gas collection pipe is pulled out to the extraction depth threshold in the opposite direction to the insertion direction, and then the exhaust gas collection pipe insertion action is performed.

2. The intelligent intubation control method for motor vehicle exhaust emission detection based on image recognition according to claim 1, characterized in that, The path cost function is obtained through the following formula: Q Value for the path d(p i ,p i+1 ) For nodes p i With nodes p i+1 The distance between them θ i+1 For nodes p i+1 Angle, θ i For nodes p i Angle, q e Based on the energy consumption of the robot arm during its movement along the motion path I This represents the total number of nodes in the motion path.

3. The intelligent intubation control method for motor vehicle exhaust emission detection based on image recognition according to claim 1, characterized in that, In step S1, the operation of identifying the exhaust pipe from the image of the rear of the car specifically involves: The edge points in the car rear image are marked to obtain a marked car rear image. Detection boxes are then placed all over the marked car rear image. Detection boxes in the same region whose center point is the minimum distance from the center point of the ground truth box are used as the corresponding classification boxes. Based on the shape, pixel distribution, and area of ​​the exhaust pipe, all classification boxes are classified. Classification boxes that are classified as exhaust pipes and whose corresponding classification probability value is greater than a probability threshold are used as exhaust pipe recognition boxes. The exhaust pipe recognition box is processed by Hough transform to obtain the center position of the exhaust pipe; the coordinates of the center position of the exhaust pipe in the image of the rear of the car are used as the two-dimensional coordinates of the exhaust pipe.

4. The intelligent intubation control method for motor vehicle exhaust emission detection based on image recognition according to claim 3, characterized in that, The specific method for obtaining edge location points is as follows: The image of the car's rear is converted to grayscale to obtain a grayscale image of the car. The weighted average of the pixels at each location point and other locations within the first neighborhood of the grayscale image of the car is taken as the new pixel value of the corresponding location point to obtain a denoised image of the car. Determine whether the pixel value at each location point in the denoised car image is greater than the edge pixel threshold; If the pixel value of the current position is not less than the first edge pixel threshold, then the current position is an edge position point; If the pixel value of the current position point is between the first edge pixel threshold and the second edge pixel threshold, and the current position point is adjacent to the determined edge position point, then the current position point is an edge position point; The threshold for the first edge pixel is greater than the threshold for the second edge pixel.

5. A vehicle exhaust emission detection intelligent intubation control system based on image recognition, used to implement the vehicle exhaust emission detection intelligent intubation control method based on image recognition as described in claim 1, characterized in that, include: The robotic arm has an exhaust gas collection pipe at its end, which is used to move the exhaust gas collection pipe according to the instructions transmitted from the control unit. A camera, mounted on a robotic arm, is used to capture images of the rear of a car. The image processing unit is used to process the rear-end image of the car captured by the camera and identify the exhaust pipe from the rear-end image of the car. The computing unit is used to convert the two-dimensional coordinates of the car rear image into the three-dimensional coordinates of the robot arm, and then perform path planning based on the position information of the exhaust pipe in the three-dimensional coordinates of the robot arm and the joint angle information of the robot arm to obtain the movement path of the robot arm. The control unit is used to move the exhaust gas collection pipe at the end of the robot arm to face the exhaust pipe according to the robot arm movement path in the computing unit. Using a segmented speed control method with a preset insertion direction, the exhaust gas collection pipe is inserted into the exhaust pipe to a preset depth threshold. And when the exhaust gas collection pipe is inserted and the insertion pressure of the exhaust gas collection pipe is greater than the pressure threshold, the exhaust gas collection pipe is pulled out to the extraction depth threshold in the opposite direction to the insertion direction, and then the exhaust gas collection pipe insertion action is performed. A pressure sensor is used to obtain the insertion pressure of the exhaust gas collection pipe.

6. The intelligent intubation control system for motor vehicle exhaust emission detection based on image recognition according to claim 5, characterized in that, There are two cameras. One camera is installed on the main post of the robot arm to capture images of the car, and the other camera is installed at the end of the robot arm to capture images of the rear of the car when the rear of the car is present in the image.

7. The intelligent intubation control system for motor vehicle exhaust emission detection based on image recognition according to claim 5, characterized in that, Also includes: A depth sensor is used to determine the depth at which the exhaust gas collection pipe is inserted into the exhaust pipe; The exhaust gas collection pipe fixing clamp is installed at the end of the robot's robotic arm and is used to fix the exhaust gas collection pipe.

Images