Single Image Super-Resolution Algorithm Based on Enhanced Generative Adversarial Network

With the advancement of technology, the processing of increasing image data becomes particularly important. Computer vision can process the hidden information behind these data. Image super-resolution aims to use a low-resolution image to generate a corresponding high-resolution image through the convolutional neural network, improving the quality of the image and the display effect. This paper proposes an Enhanced Generative Adversarial Network (EGAN), designing a new loss function, so that the generated image has more texture details than the previous method. On three benchmark datasets, Set5, Set14 and BSD100, the proposed method has a great improvement on the peak signal-to-noise ratio and structural similarity of the generated images.