Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6938302 | Journal of Visual Communication and Image Representation | 2018 | 10 Pages |
Abstract
The Anisotropic Quality Index (AQI) was previously introduced by the authors as a non-distortion-specific, no-reference image quality measure. AQI is based on measuring the variance of the expected entropy of a given image over a predefined set of orientations. In this paper, a statistical evaluation of AQI is provided using standard benchmarking parameters showing good match with observer's response. Some AQI extensions are proposed for tackling ill-posed noise types that provide fair or poor performance. A comparison with other no-reference methods has been included showing good competitiveness with many of them.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Vision and Pattern Recognition
Authors
Salvador Gabarda, Gabriel Cristóbal, Navdeep Goel,