A generalized Masi entropy based efficient multilevel thresholding method for color image segmentation

Multilevel thresholding for image segmentation is a crucial process in several applications such as feature extraction and pattern recognition. In this paper, a novel Masi entropy-based criterion for color satellite image multilevel thresholding is proposed. The proposed algorithm is based on Masi entropy which can deal with the additive/non-extensive information through the aid of a concordant entropic parameter r' which is extended in favor of multilevel based color satellite image segmentation. In addition, a comparative study between proposed Masi entropy-based color image multilevel thresholding and well known state-of-the-art entropies such as Kapur's, Renyi's and Tsallis entropy is presented. The simulation results of the proposed Masi entropy-based algorithm illustrate better performance for normal and color satellite image segmentation. Trials are conducted on various color test images to concrete the efficiency of the proposed algorithm. For segmentation purpose numerous fidelity parameters are computed such as structural similarity index (SSIM), feature similarity index (FSIM), misclassification error (ME), mean square error (MSE) and peak signal to noise ratio (PSNR).