Principal Component Analysis of LISS—III Images Using QGIS

Abstract

Processing and Classification of Remote Sensing Data is essential for various purposes such as knowing the earth’s surface and its terrain, natural resource management, natural disaster management, agriculture, soil fertility, reclamation of waste and degraded lands. Generating the Land Use Land Cover (LULC) map of the study area is the first and basic step to achieve most of the above objectives. LULC map is also useful to perform change detection studies, to know the terrain of the earth’s surface, to monitor the growth of forests, to know the aftermath of a natural disaster such as forest fire, volcano, floods. Principal Component Analysis (PCA) takes a significant role in generating the LULC map as it gives the advantage of reducing the image dimension thus reducing system and time requirements for data analysis. In this article PCA has been performed using QGIS software which is an open source software. Along with data compression, dimensionality reduction and simplicity in data analysis, PCA also supports in representing the data over the feature vector, i.e., where there is maximum variation in data.