Detection of Hotspots in Noaa/avhrr Images Using Principal Component Analysis and Information Fusion Technique

India accounts for the world’s greatest concentration of coal fires which cause several devastating environmental effects. Only Jharia Coal Field (JCF) in Jharkhand (India) contains nearly half of subsurface mine fires (hotspots) in Indian coalfields. Therefore attention is required in this direction for mapping, monitoring and detecting these hotspots. Operational satellite images can be very efficient, effective and economic tool for this purpose. Present paper deals with the potential application of operational satellite images to detect hotspots in Jharia region and proposes an algorithm for the same by employing Principal Component Analysis (PCA) and information fusion technique on NOAA/AVHRR images of Jharia region. PCA is an efficient and effective technique for finding patterns in data of high dimensions and a very powerful tool that provides a new set of images that are linear combinations of the original spectral band images. The algorithm consists of two steps: (1) application of PCA on multi-channel information along with vegetation index information obtained from NOAA/AVHRR image to obtain principal components, and (2) fusion of information obtained from principal component 1 and 2 to classify image pixels as either hotspots or non-hotspots. Results obtained with the proposed algorithm are compared with the results obtained by ground survey and a good agreement is obtained between observed and predicted hotspots. ∗ Corresponding author.