A novel approach to the classification of terrestrial drainage networks based on deep learning and preliminary results on solar system bodies

Several approaches were proposed to describe the geomorphology of drainage networks and abiotic/biotic factors determining their morphology. There is an intrinsic complexity explicit qualification morphological variations in response various types control difficulty expressing cause-effect links. Traditional methods network classification are based on manual extraction key characteristics, then applied as pattern recognition schemes. These approaches, however, have low predictive uniform ability. We present a different approach, data-driven supervised learning by images, extended also extraterrestrial cases. With deep models, phase integrated within more objective, analytical, automatic framework. Despite initial difficulties, due small number training images available, similarity between shapes samples, we obtained successful results, concluding that valid way for data exploration related fields.