A cause-based methodology for semantic analysis of deforestation using multispectral reflectance

Nowadays, deforestation activity still occurs despite having huge impact for human being. The different causes of deforestation are relatively brings different effect on nature, while forest fire and illegal logging are two major deforestation activities. In this paper, we proposed a new method to represent semantic analysis of deforestation effect based on its cause. We proposed idea to interpret reflected 'substances (material)' of deforestation area in spectrum domain into human language. The objectives of this paper are to (1) Detect deforestation area, (2) Detect the type activity that cause deforestation (whether it is logging or forest fire), (3) measuring the degree of deforestation effect for soil by considering the soil properties, (4) identify and present important nature changes occurring in soils that affect post deforestation management. Riau has been selected as the study area, where the data was acquired by using Landsat Satellite images between 2013 and 2014; where there is a big forest fire occurs. The experimental result produces new semantic analysis matrices to determining the soil condition after deforestation in different context.