An improved method for color images enhancement considering HVS

Viewing the image produces a subjective experience that corresponds with viewing the real scene. The preservation or enhancement of brightness, contrast, and color is very important for color image processing. While most previous algorithms are focusing on the luminance component, this paper presents an algorithm for color enhancement focusing on chromatic components in the compressed domain. On the basis of DC-T transform, the proposed algorithm provides different treatments on different image blocks considering the perceptual properties of Human Visual System (HVS). A simple method in reducing blocking artifacts, which is based on Weber-Fechner Law is introduced. This paper also compares the results of experiments and discusses the effects of different parameters. Experiments indicate that through choosing optimal parameter the algorithm performs better in color enhancement and blocking artifacts reduction compared to other compressed domain based approaches, such as alpha rooting. © 2012 IEEE