当前位置: 首页 > 文章 > 基于单目摄像头的车辆前方障碍物的检测研究 农业装备与车辆工程 2022,60 (1) 84-87
Position: Home > Articles > Detection of Obstacles in Front of Vehicles Based on Monocular Camera Agricultural Equipment & Vehicle Engineering 2022,60 (1) 84-87

基于单目摄像头的车辆前方障碍物的检测研究

作  者:
沈念伟;于大泳
单  位:
上海理工大学机械工程学院
关键词:
机器视觉;障碍物检测;ROI区域;图像预处理;形态学
摘  要:
针对车辆前方障碍物准确检测的需求,提出了一种基于单目摄像头的车辆前方障碍物的检测系统。通过单目摄像头对视频进行采集,获取单帧图像信息。通过在ROI区域内对道路上的障碍物进行检测,能够最大程度缩小检测范围,加快了运算效率。通过基于遗传算法的二维最大熵分割法分理出背景和障碍物,形态学算法去除其区域内的干扰物以及孤立点,最后通过填充算法和带有阈值的差分算法提取出障碍物。大量实验结果表明,通过此算法系统对道路上的障碍物检测简单、有效。
译  名:
Detection of Obstacles in Front of Vehicles Based on Monocular Camera
作  者:
Shen Nianwei;Yu Dayong;College of Mechanical Engineering,University of Shanghai for Science and Technology;
关键词:
machine vision;;obstacle detection;;ROI area;;image preprocessing;;morphological
摘  要:
Aiming at the requirement of accurate detection of vehicle obstacles in front,a detection system based on monocular camera is proposed.The video is collected by monocular camera to obtain single frame image information.Obstacles on the road are detected in the ROI region,which can minimize the detection range and accelerate the computing efficiency.The background and obstacles are sorted out by two-dimensional maximum entropy segmentation method based on genetic algorithm,and the interferences and outliers in the region are removed by morphological algorithm.Finally,the obstacles are extracted by filling algorithm and difference algorithm with threshold.A large number of experimental results show that the algorithm is simple and effective for obstacle detection on the road.

相似文章

计量
文章访问数: 11
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊