An effective technique to detect forest fire region through ANFIS with spatial data

Recently, spatial data mining plays a vital role because of its pressing need in the real world applications. Among the wide range of applications the nature disaster diagnosis is the most imperative and besides the other forest fire region mining through the spatial data is most imponderable. Forest fires are an increasing threat to the environment in both the tropical and boreal regions of the world. Hence the detection of fire regions of forest fires through the remote sensing images is most important. In this work, the forest fire region is detected through the spatial data in three phases, preprocessing phase, training phase and detection phase. Initially in the preprocessing phase, the remote sensing forest fire images are filtered through the unsharp filtering and then the image is converted to the CIEXYZ color space. Then the XYZ color spaced image is segmented through the anisotropic diffusion technique. After that the segmented regions are trained the Adaptive Neuro Fuzzy Inference System (ANFIS) in training phase. In the detection phase, the test image is tested in the trained ANFIS after the completion of preprocessing process and consequently the fire regions of the forest fire are detected. Hence the fire regions of forest image are identified and detected in an effective manner.