A comparative approach of identification and segmentation of forest fire region in high resolution satellite images

Forest, complex ecosystem, is a shelter for all living things such as plants, animals, birds and various resources such as lakes, rivers, minerals. Forests cover approximately 30% of land on the earth. As compared to forest expansion by natural or planting, the deforestation rate is always higher. Forest fire is the most notorious danger which entirely ruined the environment of the forest. These giant size fires can spread out and change the direction rapidly from its source. It is very difficult to monitor and control the fires in remote and vast areas like forest. In this research work modified fuzzy c-means clustering approach had proposed for the identification and the extraction of the region of forest fires. The experiment is conducted on both RGB and CIELab color space. The outcome of the modified fuzzy c-means clustering method had compared with K-means clustering method based on the evaluation of the image quality parameters. Experimental results had demonstrated the efficiency for the proposed approach for the detection and segmentation of forest fires in high resolution satellite images.