A fast adaptive image restoration filter for reducing block artifact in compressed images

We propose a fast adaptive image restoration filter using DCT-based block-classification for reducing block artifact in compressed images. More specifically, edge direction of each block is classified by using the DCT coefficients, and the adaptive constrained least squares (CLS) filter with corresponding direction is used for smoothing the discontinuity between blocks. For real-time application, the proposed restoration filter can be realized in the form of a truncated FIR filter, which is suitable for postprocessing reconstructed images in HDTV, DVD, or video conferencing systems.