TY - JOUR
T1 - Building-A-Nets
T2 - Robust Building Extraction from High-Resolution Remote Sensing Images with Adversarial Networks
AU - Li, Xiang
AU - Yao, Xiaojing
AU - Fang, Yi
N1 - Funding Information:
Manuscript received April 24, 2018; revised July 2, 2018; accepted August 9, 2018. Date of publication August 30, 2018; date of current version October 15, 2018. This work was supported in part by the Jiangsu Province Geographic Information Research Project (JSCHKY201720), in part by the National Science technology Support Plan Project of China under Grant 2015BAJ02B00, and in part by the China Scholarship Council (201704910704). (Corresponding author: Yi Fang.) X. Li is with the Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China, and also with the Department of Electrical and Computer Engineering, New York University, Brooklyn, NY 11201 USA (e-mail:,xl1845@nyu.edu).
Publisher Copyright:
© 2008-2012 IEEE.
PY - 2018/10
Y1 - 2018/10
N2 - With the proliferation of high-resolution remote sensing sensor and platforms, vast amounts of aerial image data are becoming easily accessed. High-resolution aerial images provide sufficient structural and texture information for image recognition while also raise new challenges for existing segmentation methods. In recent years, deep neural networks have gained much attention in remote sensing field and achieved remarkable performance for high-resolution remote sensing images segmentation. However, there still exists spatial inconsistency problems caused by independently pixelwise classification while ignoring high-order regularities. In this paper, we developed a novel deep adversarial network, named Building-A-Nets, that jointly trains a deep convolutional neural network (generator) and an adversarial discriminator network for the robust segmentation of building rooftops in remote sensing images. More specifically, the generator produces pixelwise image classification map using a fully convolutional DenseNet model, whereas the discriminator tends to enforce forms of high-order structural features learned from ground-truth label map. The generator and discriminator compete with each other in an adversarial learning process until the equivalence point is reached to produce the optimal segmentation map of building objects. Meanwhile, a soft weight coefficient is adopted to balance the operation of the pixelwise classification and high-order structural feature learning. Experimental results show that our Building-A-Net can successfully detect and rectify spatial inconsistency on aerial images while archiving superior performances compared to other state-of-the-art building extraction methods. Code is available at https://github.com/lixiang-ucas/Building-A-Nets.
AB - With the proliferation of high-resolution remote sensing sensor and platforms, vast amounts of aerial image data are becoming easily accessed. High-resolution aerial images provide sufficient structural and texture information for image recognition while also raise new challenges for existing segmentation methods. In recent years, deep neural networks have gained much attention in remote sensing field and achieved remarkable performance for high-resolution remote sensing images segmentation. However, there still exists spatial inconsistency problems caused by independently pixelwise classification while ignoring high-order regularities. In this paper, we developed a novel deep adversarial network, named Building-A-Nets, that jointly trains a deep convolutional neural network (generator) and an adversarial discriminator network for the robust segmentation of building rooftops in remote sensing images. More specifically, the generator produces pixelwise image classification map using a fully convolutional DenseNet model, whereas the discriminator tends to enforce forms of high-order structural features learned from ground-truth label map. The generator and discriminator compete with each other in an adversarial learning process until the equivalence point is reached to produce the optimal segmentation map of building objects. Meanwhile, a soft weight coefficient is adopted to balance the operation of the pixelwise classification and high-order structural feature learning. Experimental results show that our Building-A-Net can successfully detect and rectify spatial inconsistency on aerial images while archiving superior performances compared to other state-of-the-art building extraction methods. Code is available at https://github.com/lixiang-ucas/Building-A-Nets.
KW - Adversarial network
KW - building extraction
KW - fully convolutional DenseNet (FC-DenseNet)
KW - remote sensing
KW - structural feature learning
UR - http://www.scopus.com/inward/record.url?scp=85052828153&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85052828153&partnerID=8YFLogxK
U2 - 10.1109/JSTARS.2018.2865187
DO - 10.1109/JSTARS.2018.2865187
M3 - Article
AN - SCOPUS:85052828153
SN - 1939-1404
VL - 11
SP - 3680
EP - 3687
JO - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
JF - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
IS - 10
M1 - 8453267
ER -