Automatic identification of charcoal origin based on deep learning

The differentiation between the charcoal produced from (Eucalyptus) plantations and native forests is essential to control, commercialization, supervision of its production in Brazil. main contribution this study identify origin using macroscopic images Deep Learning Algorithm. We applied a Convolutional Neural Network (CNN) VGG-16 architecture, with preprocessing based on contrast enhancement data augmentation rotation over training set images. performance CNN fine-tuning 360 plantation forests. results pointed out that our method provides new perspectives origin, achieving upper 95 % mean accuracy classify for all compared strategies.