Research On Intelligent Vehicle Obstacle Detection Method Based On Image Recognition

Intelligent vehicle is a comprehensive control platform that includes functions such as environmental perception,path planning,and auxiliary driving.Obstacle detection as a key technology in intelligent vehicle autonomous navigation,has always been the focus of the relevant personnel research.Aiming at the problems in view of existing obstacle detection method,such as low adaptability and low recognition rate in the environment of lighting,multi-objective and background interference,an obstacle detection method based on restricted region growing and saturation characteristics is proposed.Then the monocular vision model is used to distance the obstacle target,and the image distance of the obstacle is further converted into the actual distance in the space.Finally,the intelligent vehicle obstacle detection system was designed.The details are as follows:(1)A new method based on restricted region growing and saturation characteristics is proposed.Firstly,the obstacle detection image containing the edge information is processed by the restricted region growing,divided the target area where the obstacle.Then the method based on the saturation characteristics is used to fine locate the obstacle target,the final result of the obstacle detection is obtained.An obstacle identification simulation software is designed,and the obstacle detection experiment is carried out on this basis,and the coordinates of the feature points on the obstacle are extracted.Finally,the obstacle detection experiment is carried out by using the method in the literature,and the validity and feasibility of the method are verified.(2)The geometric spatial coordinate transformation is used to establish the mapping relationship between the world coordinate system,the camera coordinate system,the imaging coordinate system and the image coordinate system.The camera is set up to a certain inclination angle,and a single frame still image distance model is established.The internal and external parameters and lens distortion of the camera are extracted by using the checkerboard of the checkerboard.On this basis,the single position distance measurement experiment is carried out to calculate the accurate position information of the obstacle target.Finally,In the method of the obstacle range comparison experiment.(3)The intelligent vehicle obstacle detection system was analyzed and designed,Select VS2010 as the system development platform,use C + + programming language and combined with OpenCV visual function library,in the MFC framework to build a intelligent vehicle obstacle detection system simulation platform interface.The obstacle identification and distance measurement methods are transplanted into it,and the simulation experiments of intelligent vehicle obstacle detection are carried out.Finally,the anti-collision early warning mechanism of intelligent vehicle is designed,and the anti-collision early warning experiment is carried out on this basis.The function of identification,distance measurement and anti-collision warning of the intelligent vehicle is simulated and realized.Provided the navigation information for the hardware implementation of intelligent vehicle.