<![CDATA[Image segmentation is crucial in image processing and computer vision, with multilevel thresholding (ML-ISP) offering robust solutions for complex images. However, effectively applying ML-ISP to RGB color images remains a challenge due to computational complexity and the limitations of traditional optimization algorithms, such as the Grey Wolf Optimizer (GWO). This study proposes an Inspired Grey Wolf Optimizer (IGWO) to address these issues and enhance ML-ISP for RGB color images. The performance stability of IGWO is comprehensively evaluated using three distinct objective functions: the Otsu method, the Kapur Entropy, and the M. Masi Entropy. Qualitative and quantitative analyses using PSNR, SSIM, and UQI were conducted on benchmark images. Results consistently demonstrate that IGWO, particularly with M. Masi Entropy, achieves superior segmentation quality. This research incorporates GridSearch-based hyperparameter tuning. The findings highlight the effectiveness and robustness of the proposed IGWO approach for complex ML-ISP tasks on color images.]]>
Copyrights © 2025
