<p>Land Use and Land Cover (LULC) change detection in hyperspectral images reveals a significant amount of information that may remain unexplored in traditional imaging techniques. In this research work, a variable-length non-dominated sorting genetic algorithm is introduced that considers both the spectral and spatial features of hyperspectral images. The algorithm is designed to detect and measure changes in Land Use and Land Cover over time. The efficiency of the proposed method is evaluated by comparing it with six existing approaches that address similar challenges in detecting LULC changes. Metrics such as Average Accuracy, Overall Accuracy, and the Kappa Coefficient are used to assess the accuracy and robustness of these methods. The results clearly show that the proposed technique outperforms the existing techniques. These results highlight the strength of the proposed framework in accurately and efficiently detecting LULC changes, demonstrating its strong potential to significantly improve land cover assessment methods.</p>