Application of Compromising Evolution in Multi-objective Image Error Concealment
This work addresses multi-objective optimization challenges in image enhancement, specifically for error concealment, but appears incremental as it modifies an existing genetic algorithm.
The paper tackled the problem of multi-objective optimization in image error concealment, where existing convex methods often fail due to unknown mutual preferences and lack of generative models, and proposed the Compromising Evolution Method to modify a Simple Genetic Algorithm, with simulation results demonstrating its effectiveness in solving such optimizations.
Numerous multi-objective optimization problems encounter with a number of fitness functions to be simultaneously optimized of which their mutual preferences are not inherently known. Suffering from the lack of underlying generative models, the existing convex optimization approaches may fail to derive the Pareto optimal solution for those problems in complicated domains such as image enhancement. In order to obviate such shortcomings, the Compromising Evolution Method is proposed in this report to modify the Simple Genetic Algorithm by utilizing the notion of compromise. The simulation results show the power of the proposed method solving multi-objective optimizations in a case study of image error concealment.