A MTIS method using a combined of whale and moth-flame optimization algorithms

Finding the threshold vector that gives the best performance of the image segmentation system is significant in Multi-level Thresholding Image Segmentation (MTIS) methods. Meta-Heuristic (MH) algorithms are among the techniques that can find reasonably good optimal thresholds and require reasonable computational resources. We use the combination model of the Whale Optimization Algorithm (WOA) and in conjunction with Moth-Flame Optimization (MFO) for MTIS. In MFWOA, the solutions during the exploitation phase are updated using the operators of WOA, and in the exploration phase, only the operators of MFO are used. The Inverse Otsu (IO) Function is used as Fitness Function for MFWOA. Experiments in image segmentation show that the proposed MFOWOA method is better than the compared algorithms in terms of accuracy as indicated by two performance measures: Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). It is also observed that the MFWOA algorithm is faster than WOA and slower than MFO in terms of execution time evaluation metric. In some cases, the proposed algorithm is faster than other algorithms. The results show demonstrate that the hybrid MFWOA algorithm solves MTIS problems better than both WOA and MFO algorithms and can obtain better thresholds that increase the performance of the MTIS system.