SAR Image Super-Resolution Base on Weighted Dense Connected Convolutional Network
Jianwen Yu, Wenchao Li, Zhongyu Li, Junjie Wu, Haiguang Yang, Jianyu Yang · 2020
In this paper, a weighted dense connected convolutional network(WDCCN) is proposed for SAR image super-resolution. In the network, to enhance feature propagation and the super-resolution performance, each layer will receive the output from all the previous layers in a different weight proportion. At last, the experimental results indicate that weighted dense connected convolutional network can realize SAR image super-resolution well.