No abstract

Image quality evaluation standard of information hiding algorithm mainly relied on the traditional standard such as peak signal to no ise ratio (PSNR), but PSNR does not meet people's subjective assessment. So it is necessary to set up the ob jective evaluation criteria with subjective uniformity for information hiding algorithm. In this paper, an image subjective qu ality evaluation model is proposed by structure and hue similarity (SHSIM) with human visual system (HV S), the corresponding relation between ITU-R quality and impairment scales model and objective assessment by SHSIM is set up from these three aspects as structure, color space and human visual system. The experimental results show that the image quality evaluation model has better agreement with the human's visual judgment.

Abstract. Image quality evaluation standard mainly relied on the traditional standard like peak signal to noise ratio (PSNR), but PSNR does not meet people's subjective assessment. To obtain more accurate evaluation with subjective and objective uniformity, the multi-indexes are considered. Our evaluation system includes three parts: (1) grade evaluation; (2) auxiliary evaluation; (3) reference evaluation. The value of grade evaluation is mainly determined by the grade evaluation, and minor modifications are made by the auxiliary evaluation. The reference evaluation provides the traditional objective standards so as to comparative analysis. The results of actual operation show that it is comprehensive and effective.