Sharma-Mittal Entropy and Whale Optimization Algorithm Based Multilevel Thresholding Approach for Image Segmentation

Abstract

AbstractImage thresholding is one of the most effective segmentation approaches. Multilevel thresholding approach is widely applied for segmentation in cases of complex images. In multilevel thresholding, the main goal is to search optimal threshold values for segmenting the given image into appropriate regions. In this paper, Sharma-Mittal entropy-based objective function is designed to formulate the problem of searching optimal threshold values in multi-thresholding. Here, the main issue is that the time complexity of the searching procedure increases exponentially as levels of thresholding increase. Thus, this is a NP-hard combinatorial optimization problem. To solve this problem, whale optimization algorithm (WOA) is applied for searching optimal thresholds in multilevel thresholding. WOA belongs to evolutionary computing techniques category that mimics the social behavior of humpback whales and is inspired by the bubble-net hunting strategy. Thus, in this paper, a new Sharma-Mittal entropy-based multilevel thresholding approach using WOA has been proposed. This approach is called WOA-based SMEMT approach. In the proposed approach, the optimum threshold values at different levels are searched by the minimization of objective function based on Sharma-Mittal entropy using WOA. The proposed approach is evaluated on benchmark images of standard databases, and experiment results are compared with gray wolf optimizer (GWO), teaching–learning-based optimization (TLBO), and genetic algorithm (GA) algorithms using segmented image quality evaluation parameters like peak signal to noise ratio (PSNR), uniformity, structure similarity (SSIM) index, and mean structure similarity (MSSIM) index. Results show that the proposed approach is performing better than the other approaches.KeywordsSharma-Mittal entropyWhale optimization algorithm (WOA)Multilevel thresholdingNP-hardSSIMMSSIM