An intraframe error concealment: nonlinear pattern alignment and directional interpolation

This paper presents a new intraframe error concealment algorithm designed for both image and video with nonlinear pattern alignment and directional interpolation. By this intraframe error concealment algorithm, first the nonlinear pattern alignment based on a variant version of the dynamic programming algorithm is applied to image pixels around a corrupted region. The nonlinear pattern alignment effectively provides the best aligned pixel pairs from boundary pixels around the corrupted region to dynamically capture local image textures in terms of optimized metric. Then, with the aligned pixel pairs, the directional interpolation algorithm is used to recover image pixels in the corrupted region. From computer simulation in both image and video, the proposed intraframe error concealment provides much better quality improvement of recovered image pixels comparing to that of the bilinear interpolation approach.