LULC classification for change detection analysis of remotely sensed data using machine learning-based random forest classifier

Abstract

Land Use and Land Cover (LULC) classification is critical for monitoring and managing natural resources and urban development. This study focuses on LULC classification for change detection analysis of remotely sensed data using a machine learning-based Random Forest classifier. The research aims to provide a detailed comparative analysis of LULC changes between 2010 and 2020. The Random Forest classifier was chosen for its robustness and high accuracy in handling complex datasets. The classifier achieved a classification accuracy of 86.56% for the 2010 data and 88.42% for the 2020 data, demonstrating an improvement in classification performance over the decade. The results indicate significant LULC changes, highlighting areas of urban expansion, deforestation, and agricultural transformation. These findings underscore the importance of continuous monitoring and provide valuable insights for policymakers and environmental managers. The study demonstrates the efficacy of using advanced machine-learning techniques for accurate LULC classification and change detection in remotely sensed data.