A roadside vehicle recognition method based on vision sensor

Abstract

A roadside vehicle recognition method based on visual sensors. This method aims at the low accuracy of multi-scale vehicle target recognition from the roadside perspective. A roadside vehicle recognition network based on visual sensors is built. Firstly, a residual attention module is constructed; secondly , introduce the residual attention module into the multi-scale feature fusion structure of the YOLOv3 network to form a roadside vehicle recognition network YOLO-A based on the residual attention module; then, add the anchor-free feature on the basis of the YOLO-A network Select the module to form the roadside vehicle recognition model YOLO‑AF that improves the YOLOv3 network; next, train the YOLO‑AF network; finally, use the trained YOLO‑AF network for roadside vehicle recognition. A roadside vehicle recognition method based on a visual sensor proposed by the present invention has high accuracy and good real-time performance.

Description

technical field [0001] The invention belongs to the field of deep learning and environment perception, and relates to a roadside vehicle identification method based on a visual sensor. Background technique [0002] With the rapid development of modern society and economy and the acceleration of urbanization, automobiles are playing an increasingly important role in people's production and life. With the rapid increase of vehicles, traffic has gradually become a hot issue of social development. The intelligent vehicle-road system came into being, aiming to improve the traffic conditions with the help of modern science and technology, and realize a high degree of coordination of "people-vehicle-road-cloud", which is an effective way to solve the current urban traffic problems. [0003] In the intelligent vehicle-road system conditioned on the intelligence of vehicles and roads, "smart vehicles" and "smart roads" are important dual-drive systems. A "smart car" is a smart car,...

AI Technical Summary

Problems solved by technology

[0005] In the complex traffic scene on the roadside, the vehicles are distributed from far to near, and the richness of the far and near vehicle information obtained by the roadside perception sensor is different, especially for the distant vehicles, the data resolution is low, and it is difficult to accurately identify and locate

Images