Papers › Perceptual Error Logarithm: An Efficient and Effective Analytical Method for...

Perceptual Error Logarithm: An Efficient and Effective Analytical Method for Full-Reference Image Quality Assessment

25 Apr 2025IEEE Access 2025 4archive 2025-07-28

SERGIO A. C. BEZERRA, SÉRGIO A. C. BEZERRA JÚNIOR, JOSÉ L. DE S. PIO, JOSÉ R. H. DE CARVALHO, Keiko V. O. Fonseca

In this paper, we propose a new perceptual analytical method for full-reference image assessment. Techniques used in multimedia applications often introduce distortions, which can significantly degrade image and video quality. To mitigate these distortions, image quality assessment (IQA) methods have been employed to enhance these techniques. However, efficiency and effectiveness are difficult to achieve simultaneously when designing perceptual IQA methods. To face this challenge, we introduced a new method, named the namely Perceptual Error Logarithm (PEL). Initially, luminance and chrominance maps are generated after the reduction and spatial color conversion applied to the input images. Through the use of chrominance and luminance maps, simple absolute difference maps are generated, as well as special feature maps. The special maps are derived from perceptual feature extractions applied to gradient magnitude maps. These maps are generated immediately following the application of local standard deviation filtering to the luminance maps. Finally, to calculate the PEL score, a logarithmic function is applied to the perceptual error map, which combines the simple and special maps, besides adjusts this combination based on the weight of the superpixel similarity and local energy maps. Experimental results and discussions on eight databases prove the high level of effectiveness of the proposed method, which presents, according to the application of evaluation metrics, much more consistency with subjective evaluations than all the nineteen state-of-the-art IQA methods that were compared. Moreover, PEL is highly efficient because its processing is performed in real-time.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Results from the paper archive 2025-07-28

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections