摘要
设计一种基于视频图像的行人检测系统,系统基于ARM和Linux操作系统搭建嵌入式平台,通过构建视频流服务器将网络视频流传输至PC端,利用Haar-like特征和Adaboost算法框架训练分类器,用分类器进行行人检测.详细描述系统的设计实现方案,对其中的关键点进行阐述.实验结果表明,系统性能稳定、识别精度较高.
Design a pedestrian detection system based on video image. System based on Linux OS and ARM embedded platform, by constructing a video streaming server network video streaming to the PC side, with Haar-like features and Adaboost algorithm framework for training classifiers and classifier for pedestrian detection. This paper describes in detail the design and implementation of system, and the key points discussed, experimental results show that the system is stable, high precision.
出处
《湖北大学学报(自然科学版)》
CAS
2017年第1期45-49,共5页
Journal of Hubei University:Natural Science
基金
湖北省教育厅重点科研项目(D2016004)资助