A New super-resolution restoration method with Generated Adversarial Network for underground video images in coal mines

Abstract

Abstract The computer can be used in Super-resolution reconstruction (SR) to process low-resolution images to obtain high-resolution images. Aiming at solving problems of complex underground video image acquisition environment, uneven brightness, blurred images etc, this paper adopts the idea of deep learning to perform super-resolution restoration of underground video images in coal mines, and proposes a generational confrontation network to super-resolution underground video images in coal mines. The experiment proves that Generated Adversarial Network (GAN), while being compare with Super-resolution Deep Convolutional Neural Network (SRCNN), Efficient Sub-Pixel Convolutional Neural Network (ESPCN), Deeply Recursive Convolutional Network (DRCN) the effect of GAN method is better, because it can better realize the super-resolution restoration of underground video images in coal mines and provide preliminary support for the subsequent and further application research of underground images in coal mines.