Particle Swarm Optimization based feature selection with novel fitness function for image steganalysis
Vahid Rostami, Azar Shahmoradi Khiavi · 2016
One of the most effective issues on the performance of steganalysis is feature selection that aims to select the most significant and influential feature elements. In this paper, a new method of feature selection is proposed based on optimization process of Particle Swarm Optimization (PSO) with novel Area Under Area the receiver operating characteristics Curve (AUC) measure as the fitness function to improve the performance of detecting stego images from the cover images in steganalysis problem. Due to the high convergence rate and specific search strategy of PSO, it is able to find the best feature subset and so, the performance of steganalysis will be improved. The proposed method is evaluated on Breaking Out Steganography System (BOSS) benchmark and the obtained results proves the ability of the proposed method in feature selection for steganalysis problem.