Deblocking of compressed images based on all phase biorthogonal transform using noise injection followed by image denoising

Block-based transform coding is the most popular image compression method. However, when image is highly compressed at low bit rate, noticeable blocking artifacts produce in the decoded images. Various techniques have been proposed for the reduction of blocking artifacts. This paper presents a new deblocking algorithm based on the all phase biorthogonal transform using noise injection followed by image denoising for further performance improvement. Experimental results show that better performance is achieved than the conventional algorithm in terms of PSNR and visual quality due to blocking artifacts have been reduced greatly.