Automated NDVI generation from LISS-III Images for Vegetation Monitoring in an Arid Environment: A Case Study in Jodhpur, India

Abstract

Recent advancements in remote sensing technologies have significantly improved the capability to observe, monitor, and analyse the Earth’s surface at multiple spatial and temporal scales. Satellite-based multispectral imagery has become a vital resource for environmental monitoring, land-use planning, agricultural assessment, and ecosystem health analysis. This paper presents an automated image processing algorithm developed to compute the Normalized Difference Vegetation Index (NDVI) using LISS-III multispectral data acquired over Jodhpur district, Rajasthan, India. The algorithm systematically preprocesses multiple satellite image tiles, performs mosaicking based on geographic coordinates and calculates NDVI for a user-defined region of interest. MATLAB is used as the primary computational environment to automate the entire workflow, minimizing human intervention and ensuring consistent processing. The results demonstrate the effectiveness of NDVI in distinguishing dense vegetation, sparse vegetation, and non-vegetated surfaces in an arid environment. The study highlights the importance of automated remote sensing workflows for vegetation monitoring in semi-arid and desert regions, where spatial variability and climatic extremes significantly influence vegetation distribution.