计算机技术与发展2016，Vol.26Issue(9)：87-90,4.DOI:10.3969/j.issn.1673-629X.2016.09.020
基于视频分析技术的车距测量及预警系统设计
Design of a Vehicle Distance Measurement and Early Warning System Based on Video Analysis Techniques
摘要
Abstract
Vehicle distance measuring and early warning is an important component in vehicle active safety technology. And the vision-based vehicle distance measuring and early warning system has been a research hotspot of the intelligent vehicle system and secondary safety system. In order to improve the accuracy and processing speed of vehicle distance measurement,vehicle distance and early warning system is designed and implemented based on video analysis techniques in Visual C++6. 0 and OpenCV software environment,which has functions like vehicle detection,vehicle tracking,distance measurement and early warning and so on. Haar-like features is chosen as im-age descriptions,and Adaboost algorithm is combined to train classifiers to achieve vehicles detection. A combination method of CamShift and Kalman is used to track the target vehicle,and the measurement and prediction of distance is achieved by using RBF neural network. Experiments show that the system can accurately realize the range of 1~15 m for vehicle detection and distance measurement,and it has good real-time performance.关键词
车辆检测/Haar-like特征/Adaboost算法/跟踪/RBF神经网络Key words
vehicle detection/Haar-like feature/Adaboost algorithm/tracking/RBF neural network分类
信息技术与安全科学引用本文复制引用
王传钦,曹江涛,姬晓飞..基于视频分析技术的车距测量及预警系统设计[J].计算机技术与发展,2016,26(9):87-90,4.基金项目
国家自然科学基金资助项目(61103123,61203021) （61103123,61203021）
