Satellite Image Segmentation by Using Grey Wolf Optimizer with Masi Entropy

L Bandikolla and AKM Khairuzzaman, JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING, 53, 3295-3315 (2025).

DOI: 10.1007/s12524-025-02182-3

Image segmentation is an essential technique in computer vision and image processing, offering a broad spectrum of applications, such as satellite and aerial imagery, medical imaging, object detection, agriculture, industrial inspection, document processing and traffic monitoring. Satellite image segmentation is used to detect various objects such as water bodies, roads, buildings and disaster-affected areas. This work proposes a new method for multilevel thresholding based satellite image segmentation using Masi entropy as an objective function. The threshold searching process is optimized by the grey wolf optimizer. The quality of segmented images is assessed using widely recognized evaluation metrics, including peak signal to noise ratio, mean structural similarity, feature similarity. Additionally, the entropic value is measured using Masi entropy, while computational complexity is assessed based on CPU run time. Together, these metrics efficiently offer a thorough assessment of both the segmentation quality and the algorithm's performance. The proposed method is compared with several existing methods, such as the Genetic Algorithm with Kapur's entropy, the Multilevel Thresholding-moth Flame Optimization with the Otsu method, Chaotic Darwinian Particle Swarm Optimization with the Otsu method and the Dynamic Harris Hawks with the mutation using Kapur's entropy. The comparative study shows that the proposed method can be efficiently applied in satellite image segmentation and higher level image processing and computer vision applications that require segmentation.

Return to Publications page