A NEURAL SYSTEM FOR DEFORESTATION MONITORING ON LANDSAT IMAGES OF THEAMAZON REGION

We deal with the problem of automating the interpretation of satellite images of the Amazon region for deforestation monitoring. Our approac h is based on a combination of image segmentation and classification t echniques, the latter employing a neural-network architecture that wor ks on a fuzzy model of classification. The architecture implements a r elaxation mechanism on top of a feedforward neural network, in order t o take advantage of the interrelations among neighboring image segment s. Our fuzzy, segment-based approach has numerous advantages over more traditional, pixel-based approaches employing statistical techniques. These advantages range from the possibility of treating transition an d interference phenomena in the images to the ease with which complex information related to a region's geometry, texture, and contextual se tting can be used. We report on a great variety of experiments on repr esentative portions of the Amazon region, employing neural networks tr ained by the back-propagation algorithm. The results indicate very goo d overall performance, and allow us to draw some conclusions regarding the effectiveness of the various sources of information available as input to the system. In particular, it appears that simple spectral in formation, together with textural information on a region's entropy an d correlation and simple contextual information, are effective in the classification for deforestation monitoring. It also appears that the effective incorporation of geometric information would require further investigation on possible enhancements to the system.