Blocking artifacts reduction using discrete cosine transform

Abstract

The block discrete cosine transform (BDCT) is one of the most frequently used tools for image coding. At high compression ratios, the BDCT often brings about blocking artifacts owing to coarse quantization of its coefficients. This paper presents new algorithms for reducing the blocking artifacts. The algorithms use the BDCT itself to reduce the blocking artifacts. Moreover the block positions of the BDCT are shifted from the block positions of the BDCT in the coding process. To reduce the blocking artifacts, the BDCT coefficients are suppressed properly in response to the distribution of the BDCT coefficients in the coding process. To evaluate the performance of the algorithms, the coding model is defined and the simulation experiments are made. The experimental results show the algorithms have the effects of reducing the blocking artifacts and increasing the signal-to-noise ratio (SNR).