Multilevel Thresholding Using Black Widow Optimization

The key objective of any multilevel thresholding method is to obtain the optimal thresholds on a target image for the region of interest. Traditional methods of multilevel thresholding suffer from poor accuracy and computationally expensive. Optimization algorithms are found suitable to answer these difficulties effectively. In this paper, we suggest a new multilevel thresholding algorithm using Black Widow Optimization (BWO). The paper uses the unique mating behavior concept of the black widow spider to optimize six different objective functions: Otsu’s between-class variance, Tsallis entropy, Masi entropy, Kapur’s entropy, Renyi’s entropy, and fuzzy entropy for obtaining optimal threshold values. The performance of the proposed thresholding method is evaluated by generating the segmented images of a set of standard grayscale and color images using well-known performance metrics such as PSNR, SSIM, and FSIM. The experimental outcomes show a significant improvement in the performance of the proposed method in segmenting both grayscale and color images.