In order to solve the problems of trains being prone to slope collapses,debris,perimeter branches,and floating objects intruding the boundary,a high speed detection system design scheme for objects intruding railway clearance based on three-dimensional(3D)visual reconstruction was proposed.A 3D camera was used to collect 3D point cloud data of the perimeter scene during operation,and a strong real-time embedded control processing system based on a GPU+FPGA core driver was developed.This system met the needs of massive data collection and processing for railway perimeter during operation under long-distance and large-scale cross-domain conditions.It integrated advanced geometric measurement of 3D digital point cloud and digital scene reconstruction algorithm to achieve accurate geometric ranging of key targets along the railway,real-time warnings of objects intruding railway clearance,and scene analysis and identification,so as to ensure safe train operation in different environments.
关键词
铁路/检测系统/限界侵限/三维点云/场景重构
Key words
Railway/Detection System/Objects Intruding Railway Clearance/3D Point Cloud/Scene Reconstruction
维普
万方数据