Neural network technology for image downscaling

In this paper described for the first time neural network technology developed to zoom out images scale. The developed method uses a neural structure model geometric transformations to establish the connections between frames pairs of images of high and low resolution. Adjusted so in the learning process synaptic scales allow to change the scale of images of one class without substantial loss of quality.

[1]  Ivan Izonin,et al.  Learning-based image super-resolution using weight coefficients of synaptic connections , 2015, 2015 Xth International Scientific and Technical Conference "Computer Sciences and Information Technologies" (CSIT).

[2]  D. Peleshko,et al.  Analysis of invariant moments in tasks image processing , 2011, 2011 11th International Conference The Experience of Designing and Application of CAD Systems in Microelectronics (CADSM).