Research and application of neural network approaches to solving image recognition problems

The paper investigates neural network approaches to solving number recognition problems and develops an algorithm for creating authentic datasets. In the course of work, research development datasets problem is carried out. When image problems, it advisable use technologies, but often there a lack data form full-fledged training sample during recognition. An has been developed create set artificial appropriate in networks. plates wagon numbers assumed be scope application. that forms synthetic images marked up created. result application dataset supplementing when networks field numbers.