Classification of tropical rainforest using different classification algorithm based on remote sensing imagery: A study of Gunung Basor

Remote sensing technologies are used globally to derive some of crucial spatial variable parameter such as vegetation cover. Three different classification algorithm, minimum distance classifier, Mahalanobis distance classifier and maximum likelihood algorithm was applied to classify the forest area in Gunung Basor. The study area is located in Gunung Basor, Jeli. The area is a high potential growing region for different tree species. The main objectives is to develop a forest tree recognition techniques and build a classification strategy for forest tree area segmentation. By producing classification map, accuracy for the classification can be determined. Thehighest accuracy for classification map of Gunung Basor is by using maximum likelihood algorithm with an accuracy of 82.90%. Thus, this project is importantto increase theaccuracy offorest classification by usingminimumdistance classifier, Mahalanobis distance classifier and maximum likelihood algorithm to develop a techniques for forest tree recognition based on remote sensing imagery. Hence, the result from this study represent the synergistic use of high resolution opticalimagerycanbeefficienttoimprovethecharacterizationoftropicalrainforest