In order to sufficiently utilize the motion information, the feature images of multi-frame are cumulated to improve the detection probability and reduce the false alarm rate.

为了充分利用目标的运动信息，连续多帧图像的融合特征图像进行能量累积，达到进一步提高检测概率，降低虚警率。

youdao

Current methods for identifying the video need decode the frame and have high complexity which is not suitable for real-time detection in places of mass data like gateway and multi-media server.

现有的基于视频内容的离线检测方法需要解码数据包，且匹配算法复杂度高，不适于数据量庞大的流媒体服务器或网关做实时检测。

youdao