What's wrong with mean-squared error?
Digital images and human vision
Evaluation of Two Principal Approaches to Objective Image Quality Assessment
IV '04 Proceedings of the Information Visualisation, Eighth International Conference
Image quality assessment: from error visibility to structural similarity
IEEE Transactions on Image Processing
A Statistical Evaluation of Recent Full Reference Image Quality Assessment Algorithms
IEEE Transactions on Image Processing
A novel low-bit-rate image compression algorithm
PCM'10 Proceedings of the Advances in multimedia information processing, and 11th Pacific Rim conference on Multimedia: Part II
| Hi-index | 0.00 |
Recent studies have found that adoption of structural similarity index (SSIM) was successful in reflecting human visual characteristics better compared with traditional peak signal-to-noise ratio (PSNR) metrics. However, this method shows some weaknesses when evaluating the quality of blurred images and noise images. Good quality results were hardly achieved as they do not match the human visual system (HVS) well. In this paper, we propose an improved image quality assessment algorithm based on mean-edge structural similarity (MESSIM). Edge information is considered sufficiently in image quality assessment. More specifically, the distortion metric of edge structure is assessed. The experimental results have demonstrated better consistency with the subjective perception for a large range of image types.