Thin Section Photos Processing with Neural Networks
S. Polushkin, Y. Volokitin, I. Edelman, E. Sadikhov, Y. Murzaev, O. Lokhanova and S. Budennyy
Event name: ProGREss'19
Session: Digital Transformation in O&G Exploration / Цифровая трансформация в ГРР
Publication date: 05 August 2019
Info: Extended abstract, PDF ( 442.68Kb )
Price: € 20
Neural network designed for cardiovascular diagnostics was trained to identify grains on thin section photos. About 150 thin section photos were processed. The result contains grains size and shape, mineralogical composition and pores size. This information is critical at all stages of field development from rock properties and petrophysical modelling to proper selection of drilling mud and well kill fluids.