Detecting f orest degradation patterns in Southeast Cameroon

The objective of this study was to evaluate the us e of a spectral index and a contextual classifier f or detection of forest degradation associated to selec tive logging in Southern Cameroon. This methodology, already applied in the Amazon, builds the Normalized Differ ence Fraction Index (NDFI) to enhance the forest ca nopy damage signal. A contextual classification algorith m (CCA) applied later to the NDFI image enables the separation of anthropogenic disturbance. These methods were tested in a certified forest concession ar ea of Southern Cameroon, in the Congo Basin. The results show that the NDFI is able to detect infrastructure associated to most selective logging operations in the study area. The additional CCA was able to accu rately discriminate human-caused forest degradation from natural occurrences.