Reproducible Improvement of Images Quality Through Nature Inspired Optimisation

Auteurs

Parisot O., Tamisier T.

Référence

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12983 LNCS, pp. 335-341, 2021

Description

Transparent and reproducible image processing is fundamental in science and engineering, whether to prove the efficiency of an algorithm or to prepare a new corpus of images. In this paper, we propose a Nature Inspired Optimisation algorithm based on Image Quality Assessment methods to obtain a reproducible sequence of transformations that improves the quality of a given image. Preliminary tests were realized on state-of-the-art benchmarks.

Lien

doi:10.1007/978-3-030-88207-5_33