Energy-aware JPEG image compression: A multi-objective approach

Customer satisfaction is crucially affected by energy consumption in mobile devices. One of the most energy-consuming parts an application images. This paper, first, investigates that there a correlation between and image quality as well file size. Therefore, these two can be considered proxy for consumption. In next step, we focused on proposing multi-objective strategy to enhance reduce size based quantisation table (QT) JPEG compression. To this end, have used general approaches: scalarisation Pareto-based. embed our into five algorithms, including energy-aware genetic algorithm (EnMOGA), particle swarm optimisation (EnMOPSO), differential evolution (EnMODE), evolutionary (EnMOES), pattern search (EnMOPS). Also, Pareto-based methods, non-dominated sorting (NSGA-II) reference-point-based NSGA-II (NSGA-III) are embedding scheme, EnNSGAII EnNSGAIII, presented. With proposed method, user’s preferences set before starting process generates only one solution preference, while approaches generate solutions so user select preferred after process. Experimental studies show performance baseline improved metaheuristic algorithms. particular, EnMOGA, EnMOPS, EnNSGA-II perform competitively, among others. From results, all cases comparison algorithms yields worst results. Among EnMOGA EnMOPS achieve first rank 6 7 out 13 second 5 terms objective function. EnMOES achieved fifth or Regarding shows outperforms EnNSGAIII 10 hyper-volume measure, it fails 3 cases. Furthermore, statistically verify algorithm’s effectiveness Wilcoxon-signed test. Finally, sensitivity analysis parameters provided. The source code reproducing results available in: https://github.com/SeyedJalaleddinMousavirad/MultiobjectiveJPEGImageCompression.