Image Clustering by Generative Adversarial Optimization and Advanced Clustering Criteria

Tuba, Eva; Strumberger, Ivana; Bacanin, Nebojsa; Bezdan, Timea; Tuba, Milan

Clustering is the task that has been used in numerous applications including digital image analysis and processing. Image clustering refers to the problem of segmenting image for different purposes which leads to various clustering criteria. Finding the optimal clusters represented by their centers is a hard optimization problem and it is one of the main research focuses on clustering methods. In this paper we proposed a novel generative adversarial optimization algorithm for finding the optimal cluster centers while using standard and advance clustering criteria. The proposed method was tested on seven benchmark images and results were compared with the artificial bee colony, particle swarm optimization and genetic algorithm. Based on the obtained results, the generative adversarial optimization algorithm founded better cluster centers for image clustering compared to named methods from the literature.