An Improvement Image Subjective Quality Evaluation Model Based on Just Noticeable Difference

Abstract

The research of information hiding technology is a significant aspect in information security. The limitation of the development in this research is the judgment of the image quality after applying the technology. The traditional image quality evaluation standard of information hiding algorithm such as Peak Signal to Noise Ratio (PSNR) is not meeting with the human subjective assessment. Watson Just Noticeable Difference (JND) model can be used in the perceptual adjustment of the information hiding algorithm. However, JND model gives the objective quality evaluation, so setting up the image subjective quality evaluation model by just noticeable difference with subjective uniformity for information hiding algorithm is very necessary. In this paper, an improvement image subjective quality evaluation model based on just noticeable difference with Human Visual System (HVS) is proposed, the corresponding relation between ITU-R quality and impairment scales model and objective assessment by JND model is built. The experimental results show that the improvement image quality evaluation model has better agreement with the human’s visual judgment.