Automobiles play a pivotal role in modern social life.They greatly facilitate people's travel,but at the same time,they also cause social problems such as environmental pollution and traffic congestion.In order to meet these challenges,domestic and foreign scholars are actively engaged in the research of driverless technology.Among them,vehicle and pedestrian detection technology is a key link in driverless technology.However,given the diversity of vehicle and pedestrian characteristics,it has become extremely difficult to detect using a single feature extraction and classification method.However,the deep learning target detection scheme has successfully overcome this problem with its complex neural network structure,so it has attracted the attention of scholars at home and abroad.This paper aims to provide useful reference and inspiration for the research and development and innovation of vehicle-mounted vision sensors by in-depth analysis of target detection based on deep learning and multi-target tracking based on SORT.
vehicle-mounted vision sensorsmulti-target detectiontracking technologydriverless drivingvehicle and pedestrian detection
万方数据
维普
