基于图像分割的视频差错复原算法

人们可能更多的注意感兴趣的区域和倾向于对背景的改变不太敏感。根据人眼的这一视觉特性；论文提出了基于图像分割的视频差错复原算法，并实现了支持该算法的H．264编解码器。该算法利用；灵活宏块排序技术对图像帧进行分割，将图像帧分割为前景子图像和背景子图像。与背景子图像相比，对前景子图像按更高图像质量的要求进行编码，并且对对前景子图像的传输进行更有效的保护。实验结果表明，论文提出的基于图像分割的差错复原算法具有良好的容错性能，重建图像的主客观质量良好。

Error Resilient Scheme Based on Image Segmentation

In general, people think a lot of on foreground of picture, when they look at the picture. In this dessertation, we introduce a image segmentation based error resilience algorithm. The video frame is divided to foreground sub-picture and background sub-picture with FMO tool, then we do some fine code to foreground sub-picture, and implement transmission error control by forward error control packet to coded data of foreground sub-picture. Simulation results show that the overall subjective quality is considerably improved compared to the conventional coding schemes.