Research On Car Panorama Assistance System With Positioning Function

With the continuous development of information science,computational vision,and smart cars,the car's panoramic auxiliary system with display and positioning functions has been become a research hotspot for researchers in related fields.The function of panoramically visualizing the information around the car body,effectively eliminating the visual blind spot,helping the driver to understand the road condition information in a timely manner,and improving driving safety,the car panoramic assistance system has been favored by the majority of drivers.However,the current automotive panoramic display system on the market has poor real-time splicing,unclear visual effects,and lack of detection and positioning functions for moving targets(eg,position,speed,and direction).Therefore,this article will focus on the above issues,research a vehicle with a moving object detection and positioning function of the panoramic display system,the system can achieve rapid image stitching,moving target feature extraction and the target's position,speed and direction of the accurate estimation The specific research content is as follows:Firstly,a novel image mosaic method of SURF car panoramic assistant system based on wavelet transform is studied.The method uses wavelet gradient vector to extract the feature points of the image coincidence area,and achieves fast acquisition of feature points under low-frequency images.At the same time,it further uses the unidirectional matching and direction consistency of the image feature point constraints to effectively eliminate false matching point pairs,thereby improving the feature points.Matching accuracy and real-time.Secondly,this paper discusses a method of detecting the moving target of the car's panoramic assistant system based on Vi Be and five-frame difference algorithm.This method is based on the traditional Vi Be algorithm.There are ghosting and shadow problems in the detection results.The introduced five-frame difference algorithm eliminates light interference,suppresses ghosting and shadowing of moving targets,and greatly improves the detection accuracy of moving targets.Finally,the research on the moving target positioning technology of the car's panoramic assistant system based on the monocular vision.Through the introduction of a median filter algorithm to preprocess the detected moving target image,and using Canny algorithm to perform edge detection on the moving target image,based on this,the monocular visual target localization algorithm is used to move the 2D image information of the moving target.The 3D coordinate information is converted into a world coordinate system,and the Kalman filter is used to track the moving target so as to accurately estimate the position,speed,and direction of the moving target.In summary,the study of this paper has important theoretical and practical application value for the development of smart car technology and the promotion of the panoramic assistant system.