Iterative blocking artefact reduction based on local contrast information

Block based transform coding is the most popular technique for image and video compression. However, images compressed using block based algorithms exhibit highly visible blocking artifacts. We propose an iterative algorithm to reduce these artifacts using a map computed by summation of the horizontal and vertical profiles of the gradient vector magnitude. This map is then used as an input to a recursive filter to reduce the blocking effect. The results of the proposed method have been compared to some recent and efficient deblocking algorithms. Experimental results show that the proposed method is simple and yields a subjective quality that is nearly free of blocking artifacts.