Research on Remote Sensing Image Restoration from Low Resolution to High Resolution Based on Deep Learning
Download as PDF
DOI: 10.25236/AISCT.2019.065
Author(s)
Cong Gao, Yang Zhao, Chuanchang Si
Corresponding Author
Cong Gao
Abstract
Image super-resolution reconstruction is an image processing technique that processes and analyzes a low-resolution image or a series of images by computer to recover the desired high-resolution image. In this paper, the deep convolutional network is the main research object, and the improved algorithm is proposed for its deficiency, which improves the reconstruction performance and robustness of the algorithm. Aiming at the shortcomings of poor recovery and poor image detail, combined with the strong learning ability of deep network, deep neural network is introduced based on traditional sparse coding algorithm, and a better SR image method is found. Good experimental results were obtained on the dataset. The experimental results on the dataset show that the improved combination method can effectively improve the image reconstruction quality.
Keywords
Deep learning; Image resolution; Image reconstruction