Single Image Super-resolution Method for Electrical Equipment Images Based on Deep Learning

Image super-resolution (SR) has achieved wide interest in the field of computer vision since it was proposed. Single image super-resolution (SISR) refers to obtaining a corresponding high-resolution (HR) image from a low-resolution image (LR). High-resolution images have more information and thus help to improve the training accuracy of neural network models. High-resolution images are often lacking in electrical engineering due to the limitation of sensor and optical device manufacturing process and cost. Therefore, it is of great practical importance to obtain HR images by super-resolution techniques. In this paper, we propose a novel approach based on generative adversarial network (GAN). Our proposed method is better than existing methods with respect to accuracy, and the visual improvement in our performance is evident.