摘要
对形状信息的单独编码是MPEG-4编码标准特有的。在容易受错误干扰的通信网络大环境中,对形状信息的错误进行隐藏可以有效避免错误在视频帧之间传播。本算法首先定义了视频对象形状信息丢失区域的4个矩形邻域,接着在这些邻域中检测与轮廓边界有接触的线,并将它们放在几何坐标系下进行分析。然后以一个优化准则为标准将这些线进行配对,保证目标轮廓的平滑性。最终每一组线都经过反复推断和纠正直到形成一个闭合的轮廓曲线。实验结果表明,针对不同的错误率条件下的网络,本算法均能将其错误进行较好的隐藏,在算法性能上与其他形状隐藏方法相比取得了更加鲁棒的效果。
The encoding of shape information is a distinguishing feature of MPEG-4. In error prone communication networks, it is important and efficient to conceal shape errors spatially, so as to avoid propagation of errors in the video frames. In this paper, four rectangular neighbors of the missing area are firstly defined in the algorithm, and then the lines that are intersected with the borders are detected in the neighbors and redefined in a geometric coordinate system. Match them into pairs under some optimization objective, then they connect in the missing area smoothly. Each group of lines will be extrapolated and corrected recursively until there is a closed curve. The proposed algorithm could conceal errors well enough under various error rates, and is more robust than other shape error concealment methods in the algorithm performance.
出处
《燕山大学学报》
CAS
2009年第6期490-494,共5页
Journal of Yanshan University
基金
河北省教育厅自然科学基金项目(2003114)