A Review on Satellite Image Segmentation Using Metaheuristic Optimization Techniques

Image segmentation is essential in digital image processing applications. Multilevel thresholding is a popular technique for image segmentation. An image can be divide into multiple classes. In this paper, the image quality metrics such as peak signal-to-noise ratio (PSNR), mean squared error (MSE), mean structural similarity (SSIM), feature similarity (FSIM) are compared using metaheuristic algorithms. The CPU time of different metaheuristics is also compared to determine the time complexity of the techniques. The Otsu’s method, Kapur’s entropy, Tsallis entropy, and Masi entropy are used as objective functions. Some of the metaheuristic algorithms, adaptive Cuckoo search algorithm (ACS), chaotic coronavirus optimization algorithm (COVIDOA), dynamic quantum-inspired genetic algorithm (DQGA), a modified artificial bee colony algorithm (MABC), dynamic Harris hawks with mutation mechanism (DHHO/M), and particle swarm optimization algorithms (PSO) are compared. As per the comparison of the quality metrics, DHHO/M with Otsu was more efficient in average PSNR. CS with Tsallis provides best value in average MSE, ACS with Otsu gives best results in average SSIM, ACS with Tsallis is more efficient in average FSIM, and MABC with Kapur’s combination was more efficient in average CPU time.