Convolutional neural network for noise reduction to improve the quality of images obtained using active-pulse television measuring systems
A Novgorodov, N. Borodina, Mikhail I. Kuryachiy · Journal of Physics Conference Series · 2022
Abstract The paper presents a software implementation of an image recovery algorithm using a convolutional neural network. An example of using the developed algorithm in the context of noise reduction of images obtained using active-pulse television measuring systems is presented. The result obtained with a convolutional neural network is better than classical filtering algorithms (PSNR higher by 2.01 dB).