摘要
智能视频监控由于其智能性、实时性和高检测率成为交通和安防领域的热点研究方向,其中运动目标实时检测是智能视频监控中的一个基本和重要的部分。本文提出一种针对智能视频监控的运动目标检测新方法:包括视频帧的小波变换、YCb Cr通道分离、目标检测与融合等环节。与传统的检测方法相比,该方法检测运动目标速度较快,精确度较高,且同时还能抑制弱的阴影。
With its intelligence, real-time and high detection rate, intelligent video surveillance becomes a hot research direction in transport and security fields, where real-time target detection is a basic and important aspect. A new method of target detection for intelligent video surveillance is proposed, which is composed of three steps： Discrete Wavelet Transform of video frames, channel separation in YCb Cr space and extraction function based target detection. The experimental results have shown that it is more fast and accurate in target detection for intelligent video surveillance, moreover it can suppress weak shadow in the same time.
出处
《电子设计工程》
2014年第20期171-174,共4页
Electronic Design Engineering
