Application of Multi-Strategy Improved Sand Cat Swarm Optimization Algorithm in Otsu Multi-Threshold Image Segmentation
Canwei Chen · 2025
Traditional multi-threshold image segmentation methods often face challenges such as low segmentation accuracy and slow convergence speed. The application of intelligent optimization algorithms can significantly enhance the efficiency and precision of multi-threshold image segmentation. This paper proposes a Multi-Strategy Improved Sand Cat Swarm Optimization (MISCSO) algorithm. The algorithm introduces a variable spiral flight strategy during the exploration phase to update positions, improving global search capability and avoiding local optima. A Gaussian random walk strategy is incorporated in the exploitation phase to update positions, enabling enhanced local exploitation and refined solution accuracy through expanded search range via random perturbations. Additionally, a crossbar strategy is employed to refine the global optimum and the population. The MISCSO algorithm is applied to the Otsu multi-threshold image segmentation task, and experiments are conducted on four test images. Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) are used to evaluate and compare the results. Experimental outcomes demonstrate that MISCSO exhibits excellent convergence speed and segmentation accuracy, proving its effectiveness in the field of multi-threshold image segmentation.