DCT변환 영상에서의 에지정보를 이용한 디블록킹 알고리듬

Abstract

For low bit rate multimedia communication, the original image is compressed by block based DCT coding. One of the major drawbacks of this coding scheme is that they may result to visible artifacts in block boundaries. In this thesis, a new algorithm is proposed based on two separate modes in terms of number of edge pixels around the block boundary. The number of edge pixels is estimated by applying the Robert’s edge filter. In each mode, proper filtering operation is carried out in both horizontal and vertical direction. The first mode is associated with smooth region. Since flat regions are more sensitive to human visual system, strong filtering operation is applied. In second mode an adaptive low pass filter is applied. This filter is based on pixel gradient around the block boundaries. Artifacts are reduced without introducing undesired distortion while preserving the original edge of image. Although the proposed approach is simple and operates on spatial domain, the experimental results show that it improves both the subjective and objective quality of images with various features.