低频干扰下无线多媒体网络中不良视频检测算法

当前无线多媒体网络容易受到低频干扰,传统不良视频检测算法无法有效去除干扰,导致检测结果不可靠。为此,提出一种新的无线多媒体网络中不良视频检测算法。为了使运动矢量平滑,通过向量中值滤波技术对其施行处理。为了避免因摄像头位置变动造成的干扰,采用平均值法对视频进行进一步处理。在无线多媒体网络视频序列时间轴中,将局部光问题转换成关键帧识别问题。在获得关键帧的基础之上,通过HIS模型快速准确地实现肤色部分的提取,判断检测视频是否为不良视频。经实验验证,所提算法检测精度高、结果可靠,且开销小。

Detection of bad video in wireless multimedia networks under low frequency interference

Nowadays, wireless multimedia networks are vulnerable to low frequency interference, and the traditional bad video detection algorithms can not effectively remove interference, resulting in unreliable detection results. Therefore, a new method of bad video detection in wireless multimedia networks is proposed. In order to smooth motion vectors, vector median filtering is applied to process them. In order to avoid the interference caused by the change of camera position, the average value method is adopted to further process the video. In the time sequence of video sequences of wireless multimedia networks, the problem of local light is converted into key frame recognition problem. The foreground area is picked out by the corresponding key frames, and the color of the interference recognition in the background is removed. Color recognition is applied to the foreground area, and the video content is analyzed by using the key frame recognition results to determine whether the detected video is bad video. The experimental results show that the proposed algorithm has high detection accuracy, reliable results and low overhead.