Post-Processing Techniques Based On Video Decompression

Compressed image post-processing technique refers to process the compressed image which shows all kinds of artifacts due to the compression aiming at improving the quality of the image. This paper studies the post-processing techniques for still compressed image and video image. All the post-processing techniques at present can not achieve good performance either using image enhancement or using image restoration. So this paper pays emphasis on the research of post-processing at low bit-rate.This paper processes the compressed image and video images based on the block discrete cosine transform (BDCT). A major drawback of block discrete cosine transform is that the decoded images,especially at very low bit rates, exhibit highly noticeable blocking artifacts near the block boundaries. The paper propose a method to remove these artifacts without smoothing images and without loosing there perceptual features. In the foundation of the deeper research on the existing methods the paper mainly research on the following aspects in this paper:(1)A blocking reduction algorithm in BDCT domain based total variation algorithm. The algorithm principle is to minimize Total Variation with a gradient declining algorithm, and then implements Total Variation adjustment to reduce blocking artifacts. The experimental results show the affectivity of the method.(2)A blocking reduction algorithm in BDCT domain based weighted Total Variation algorithm. According to the characteristics of the image encoding method, the algorithm changes TV to weighted Total Variation. In this paper,the method for combining weighted Total Variation with DT-CWT threshold is propose. By using the properties of weighted-Total Variation with DT-CWT threshold, this method can speed up the convergence rate and reduce the number of iterations. (3)A blocking reduction algorithm based on Three-dimensional Dual-Tree Complex Wavelet Transform is applied to remove the blocking artifacts of video images. Currently the denoising algorithm's study focused on single frame images and less focused on image sequences. The algorithm based on time and spatial domains is that video image sequences regard as a three-dimensional array, then use 3D DT-CWT threshold to remove the blocking artifacts. The method effectively improves the quality of the recovered video image.