Single-frame Noise2Noise: method of training a neural network without using reference data for video sequence image enhancement

摘要

A method of training neural networks for image enhancement without using reference data is proposed based on the assumption of similarity of signals and independence of noise components in spatially proximal image pixels. This approach enables the formation of the training dataset from each frame of a video sequence by decimation into even and odd rows and columns. The training of image restoration is possible considering the markers of dynamic properties of objects in the image. The efficiency and limitations of the proposed method are studied. Its performance is evaluated using a database of images obtained at a low light intensity. (C) 2021 Optical Society of America