Dual Features Extraction Network for Image Super-Resolution

With the development of deep convolutional neural network, recent research on single image super-resolution (SISR) has achieved great achievements. In particular, networks, which fully utilize features, achieve a better performance. this paper, we propose an dual features extraction network (SRDFN). Our method uses blocks (DFBs) to extract and combine low-resolution with less noise but detail, high-resolution more detail noise. The output DFB contains advantages low- Moreover, due that number channels can be set by weighting accuracy against size model, SRDFN designed according actual situation. experimental results demonstrate proposed performs well in comparison state-of-the-art methods.