A Comparison between Neural Networks and Maximum Likelihood Remotely Sensed Data Classifiers to Detect Tropical Rain Logged-over Forest in Indonesia

Selective logging has been applied in the Indonesian tropical rain forest since the 1960. This has resulted thousands of hectare logged-over forest. In Labanan, Berau, East Kalimantan, selective logging will enter the second rotation in 2010. A comprehensive analysis on the logged over forest condition should be made before harvesting the logged over forest. One aspect that should be considered is the forest structure. The objective of this study is to compare between two classification techniques (Maximum Likelihood and Neural Network Classifiers) in characterizing the condition of logged over and unlogged tropical rain forest using satellite remotely sensed data namely: Landsat-7 ETM, JERS-1 SAR, ERS-2 SAR and Radarsat-1 SAR images. The results indicated a significant difference in structure condition between logged over and unlogged forest. The canopy closure, stem density, and basal area of logged over forest are 84%, 511 trees/hectare (ha), and 26 m/ha, respectively. The corresponding results for the unlogged forest are 90%, 583 trees/ha, and 32 m/ha, respectively. The use of neural networks classifier is found to improve the accuracy of classification result as compared to ma ximum likelihood classifier. Moreover, neural networks classifier can classify two logged over forest classes with significant difference in stem density and basal area per hectare .