ROA-CS: A Hybrid Algorithm for Image Brightness Enhancement

Abstract

Meta-heuristic optimization algorithms have been commonly applied to solve image-related problems for decades. In this paper, a hybrid algorithm of Rafflesia optimization algorithm and cuckoo optimization algorithm (ROA-CS) is proposed. Since ROA has a powerful exploration capability and CS has a powerful exploitation capability. With the advantages of two algorithms, the ROA-CS algorithm is proposed. The two algorithms use Taguchi theory as the communication strategy, and the performance of the algorithms is further improved. The proposed ROA-CS algorithm is tested on 30 benchmark functions of CEC2017, and the improved algorithm outperforms the current state-of-the-art 9 algorithms. Finally, the ROA-CS algorithm is applied to image brightness enhancement to maximize the fitness function by outputting a set of optimal parameters to improve the image’s entropy, standard deviation, and edge details. The experimental outcomes show that the algorithm is effective for enhancing image brightness. The proposed ROA-CS algorithm outperforms other image enhancement methods to a large extent in terms of visual effectiveness and quantitative image quality assessment.