A new spectral index for vegetation extraction using satellite data

Abstract

Accurate vegetation extraction is essential for environmental monitoring, agricultural management, and ecosystem analysis applications. Traditional spectral indices, however, often suffer from reduced accuracy and sensitivity under varying environmental conditions. This research aimed to address these limitations by introducing a novel spectral index, "A New Vegetation Index" (ANVI), specifically designed for enhanced vegetation extraction for satellite imagery. ANVI leverages multiple spectral bands, including near-infrared (NIR), shortwave infrared (SWIR), and visible bands, commonly available in remote sensing data. The research involves implementing ANVI within a remote sensing framework and conducting a comparative analysis against established indices across diverse geographic regions. Key metrics such as accuracy, resilience to atmospheric disturbances, and sensitivity to soil background influences were evaluated. The results demonstrated that ANVI achieved a superior overall accuracy of 97.08% in vegetation classification and greater robustness against atmospheric noise than conventional indices. Furthermore, ANVI reduced soil background influence by significantly improving its performance under complex environmental conditions. This research highlights the novelty of ANVI as a computationally efficient and reliable tool for large-scale vegetation monitoring, offering enhanced precision and adaptability for diverse applications in remote sensing and ecological management.