Development of Spectral Signature of Land Cover and Feature Extraction using Artificial Neural Network Model

The remote sensing (RS) imagery is important to the development of the spectral signature of mango orchards, vegetation, and other land-use features, and to geospatial feature extraction using artificial neural networks (ANNs). The geospatial information is useful for monitoring vegetation growth, urban development, and land-use / land-cover (LU/LC) change detection. The objective of this study to develop a spectral signature and feature extraction of land-use classes using the multi-temporal and multi-spectral (MTMS) Landsat imagery dataset. The imagery dataset has obtained three images from various sensors of the Landsat satellite system from the years 2003 to 2017. The pre-processing of the imagery is crucial for geospatial feature extraction and analysis of land-use features. The vegetation index (VI) is used in this study to monitor the health and growth of orchards, vegetation, and crop. The resulting accuracy of classification using ANNs method for different years (2017, 2010, and 2003) are 90.10%, 75.75%, and 78.37%. The results of the presented study indicated that significant changes have occurred in the study region, which has affected the environment and human activities. The information of LU/LC's situation in the region will help the urban planners and decision-makers to plan for effectively managing future LU/LC change.