Research On Recognition And Location Extraction Of Train Air Duct Based On Machine Vision

With the rapid development of my country’s modern railway system network,people have put forward higher requirements for the safety assurance of railways and related systems.The train air duct is a pipe that connects the adjacent train cars.Its function is to transport the pressurized air of the brake to each car.It is related to the performance of the entire train braking system.When disassembling or overhauling the train,it is necessary to completely disassemble the air duct of the train.At present,the disassembly of the air duct of the train adopts manual methods,which has the problems of high labor intensity,low operation efficiency and certain danger.Aiming at this characteristic,using a mobile robot equipped with a vision system for automatic removal of air ducts instead of manual operations to achieve the removal of air ducts can effectively solve the problems existing in manual operations and have a significant impact on the development of railway operation technology.important meaning.This article first analyzes the camera imaging model in the vision system,establishes the relevant coordinate system,and determines its related conversion relationship.Then,using Zhang Zhengyou calibration method to calibrate the camera used in the system,and derive its principle.Finally,the internal and external parameters and distortion coefficients of the camera were obtained by the calibration experiment,which laid the foundation for the smooth development of the follow-up research.Aiming at the problem of self-positioning between two carriages when the robot moving linearly at the bottom of the train recognizes,locates and extracts the air duct of the train,a self-positioning method based on the motion estimation principle of monocular visual odometer is proposed.According to the characteristics of the monocular camera shooting vertically upwards at the bottom of the train,the image color is darker,and when the camera is about to reach between the two carriages,there will be obvious brightness change areas in the image.Select the key frame image and extract it through image processing.Obtain the target area information in the image,estimate its own motion according to the proposed motion estimation algorithm,and verify the feasibility of the method through the construction of an experimental platform.After the mobile robot reaches between the two cars smoothly,in order to realize the spatial positioning of the train duct joints,an indirect recognition and positioning method based on the binocular vision system is proposed.According to the difficulty of identifying the joints of the air duct of the train,based on the color feature,first realize the recognition of the metal circle at the proximal end of the air duct in the HSV space,and extract its two-dimensional centroid coordinates;then according to the recognition and positioning,a certain real-time and accuracy is required.The problem is to use the combined stereo matching algorithm to reconstruct the three-dimensional coordinates of the proximal metal ring of the duct;according to the fixed positional relationship between the duct joint and the proximal metal circle,the three-dimensional coordinates of the pipe joint are obtained through spatial coordinate conversion.The feasibility of the method is also verified experimentally through the construction of an experimental platform.Finally,in order to complete the task of removing the air duct of the train by the mobile robot,it is necessary to design and analyze the structure of the robot.First,analyze the structure of the air duct itself,design a suitable mechanism to realize the extraction of the train air duct.In the design of the mobile mechanism,a crawler-type mobile mechanism is used to achieve the task of moving the mobile robot at the bottom of the train.It is used in the design of the extraction mechanism.The four-degree-of-freedom mechanism realizes the task of disassembling the air duct of the train,and then integrates the design of the whole machine to determine that its structural size meets the requirements,and finally conducts kinematics analysis on the dismantling mechanism,and conducts statics analysis on the key parts of the whole machine.Determine the rationality of the entire organization.