Abstract | ||
---|---|---|
In this paper, a Generative Adversarial Network(GAN) is proposed for data augmentation of remote sensing images abstracted from Jiangsu province in China, i.e., D-sGAN(Deeply-supervised GAN). At First, to modulate the layer activations, a down-sampling scheme is designed based on the segmentation map. Then, the architecture of the generator is UNet++ with the proposed down-sampling module. Next, the generator of this net is deeply supervised by the discriminator using deep Convolutional Neural Network(CNN). This paper further proved that the proposed down-sampling module and the dense connection characteristics of UNet++ are significantly beneficial to the retention of semantic information of remote sensing images. Numerical results demonstrated that the images generated by D-sGAN could be used to improve accuracy of the segmentation network, with a better Fully Convolutional Networks Score(FCN-Score) compared to the GoGAN, SimGAN and CycleGAN models. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/IGARSS39084.2020.9324263 | IGARSS 2020 - 2020 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM |
Keywords | DocType | Citations |
data augmentation, GAN, deep supervision, down-sampling | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ning Lv | 1 | 31 | 11.32 |
Hongxiang Ma | 2 | 0 | 0.68 |
Chen Chen | 3 | 46 | 6.93 |
Qingqi Pei | 4 | 0 | 2.03 |
Yang Zhou | 5 | 0 | 0.34 |
Fenglin Xiao | 6 | 0 | 0.34 |
Ji Li | 7 | 0 | 0.34 |