The need for data mining applications in describing, explaining and forecasting spatial patterns has been on a steady increase owing to the huge rise in the number of civilian satellite repositories and the efficient utilization of remotely sensed earth observation data for the study of earth system. Fire is one of the major causes of surface change and happens in the mass of vegetation zones across the world. Forest fires are key ecological threats that lead to deterioration of economy and environment besides endangering human lives. The motivation behind this paper is to obtain beneficial information from spatial data and use the same in the determination of spots at the risk of forest fire by utilizing data mining and artificial intelligence techniques. In this paper we have proposed a novel approach to detect the forest fire automatically from the spatial data corresponding to forest regions with the aid of clustering and fuzzy logic. Initially, the digital satellite images are converted into CIELab Color Space and clustering is performed to identify the regions showing hotspots of fire. A fuzzy set is formed with the color space values of the segmented regions which are followed by the derivation of fuzzy rules on basis of fuzzy logic reasoning for the detection of forest fires. The proposed system has been evaluated with the help of publicly available spatial data corresponding to forest regions.

Spatial data mining, Remote Sensing, Forest Fire Detection, Clustering, K- means clustering, CIELab Color Space, Fuzzy logic, Fuzzy set, Fuzzy rules