A new wavelet based deblocking algorithm for compressed images
Feng Gao, Xiaokun Li, William G. Wee · 2003
The paper addresses the problem of the deblocking of JPEG compressed images. A new efficient approach based on wavelet theory is introduced. The main idea is to use the dyadic wavelet to transform decoded blocked images into time-frequency space, then use an optimal interpolation technique to smooth high frequencies around the boundaries between adjacent blocks. Based on this idea, a new deblocking algorithm is derived. This algorithm is simple and easily implemented. Experiment shows that the algorithm improves not only visual quality, but also the peak signal-to-noise ratio (PSNR) of images.