Research And Implementation Of Panoramic View System On Vehicle

With the rapid increase in the number of domestic car used,there are more and more traffic problems.The most frequent occurrence of the accident is the car cut rub accident by the lane congestion and low-speed parking,and the cause of the accident is largely due to the driver driving a vehicle where is a lot of blind spots.This kind of accident is largely due to the fact that the driver has a large number of blind spots when driving the vehicle.The Panoramic Vision System is designed to provide the driver with a top-down panoramic view of the vehicle from top to bottom to help the driver eliminate the blind spot s and reduce the incidence of such traffic accidents.In this paper,we focus on the panoramic vision system of the vehicle,introduce the imaging principle and distortion correction method of the fisheye camera,and design the panoramic image processing algorithm for the panoramic vision system.The method of accelerating the algorithm is proposed and verified on the embedded platform.Firstly,the imaging principle of the ideal camera and the fisheye camera is introduced,and the difference and connection between them are analyzed.The distortion correction algorithm and the image interpolation principle of the fisheye camera are introduced.The method of Zhang Zhengyou plane calibration is used to obtain the fisheye camera's internal matrix and the distortion coefficient,and successfully achieve the distortion correction of the fisheye picture.Secondly,the basic principle of image perspective transformation is introduced.On this basis,the optimal perspective transformation matrix method based on least squares method is designed.Then,the limitation of traditional image stitching method based on invariant feature is analyzed,and propose the image stitching method based on checkerboard registration.Finally,the image fusion algorithm is used to eliminate the splicing gap of the image stitching.Finally,the linux operating system environment is built on the I.MX6 embedded platform and the opencv was ported to the embedded system.Then,the method of accelerating the algorithm based on the idea of mapping i s designed on the real vehicle,and the algorithm is accelerated by the multi-thread technique,and the experiment is verified on the embedded platform.The experimental results show that the method proposed in this paper has successfully realized the panoramic vision system,and achieved the intended purpose,but the quality of the panoramic pictures is to be further improved.