Figure 14. Results of WorldView-3 images. The first row
is located in Kalgoorlie, the second row is located in São
Paulo, the third row is located in Madrid, and the fourth
row is located in New Delhi. The image on the right is the
enlarged result of the left white/black circle.
state-of-the-art algorithms. In addition, it is worth noting that
the algorithm processes a 10 240×10 240 pixel image in less
than 10 seconds and in only 3 seconds for the 2048×2048 pix-
el image. The results on the WorldView-3 images proved the
expansibility of the proposed algorithm. The disadvantage of
the algorithm is that the boundaries of the extracted built-up
area are not precise enough, and finding a quick edge refine-
ment method is our future research direction. In the future, we
will try to extend our adopted
LMB-CNN
to more remote sens-
ing image scene understanding tasks, such as remote sensing
image scene classification (Li
et al.
2016b) and retrieval (Li
et
al.
2018a, 2018c, 2018d). In addition, we will also test
FCN
on
more challenging tasks, such as dim object detection (Li and
Zhang 2018), multi-source object detection (Tan
et al.
2018a),
and multi-class object detection (Li
et al.
2018b).
Acknowledgments
This work was supported by the National Key Research and
Development Program of China under grant 2018YFB0505003,
the National Natural Science Foundation of China under
grant 41601352, the China Postdoctoral Science Foundation
under grants 2016M590716 and 2017T100581, and the Hubei
Provincial Natural Science Foundation of China under grant
2018CFB501. The authors would like to thank the anonymous
reviewers for their insightful and constructive comments,
which significantly improved the quality of this article.
References
Benediktsson, J. A., M. Pesaresi and K. Amason. 2003. Classification
and feature extraction for remote sensing images from
urban areas based on morphological transformations.
IEEE
Transactions on Geoscience and Remote Sensing
41:1940–1949.
Bouzekri, S., A. A. Lasbet and A. Lachehab. 2015. A new spectral
index for extraction of built-up area using Landsat-8 data.
International Journal of Remote Sensing
43:867–873.
Chen, L., Y. Zhu, G. Papandreou, F. Schroff and H. Adam. 2018.
Encoder-decoder with atrous separable convolution for semantic
image segmentation. Pages 801–808 in
Proceedings of European
Conference on Computer Vision
, held in Munich, Germany.
Chen, Y., K. Qin, H. Jiang, T. Wu and Y. Zhang. 2016. Built-up area
extraction using data field from high-resol
Pages 437–440 in
Proceedings of IEEE Inte
and Remote Sensing Symposium
, held in
Clevert, D. A., T. Unterthiner and S. Hochreiter.
accurate deep network learning by exponential linear units
(ELUs).
Computer Science
1511.
Dou, Y., Z. Liu, C. He and H. Yue. 2017. Urban land extraction using
VIIRS nighttime light data: an evaluation of three popular
methods.
Remote Sensing
9:175.
Gao, D. L., R. Zhang and D. X. Xue. 2017. Improved fully
convolutional network for the detection of built-up areas in
high resolution SAR images. Pages 611–620 in
Proceedings of
9th International Conference on Image and Graphics
, held in
Shanghai, China.
Ghosh, A., M. Ehrlich, S. Shah. 2018 Stacked U-nets for ground
material segmentation in remote sensing imagery. Pages 257–261
in
Proceedings of the IEEE Conference on Computer Vision and
Pattern Recognition Workshops
, held in Salt Lake City, Utah.
Guo, X., X. Huang and L. Zhang. 2014. Three-dimensional wavelet
texture feature extraction and classification for multi/
hyperspectral imagery.
IEEE Geoscience and Remote Sensing
Letters
11:2183–2187.
He, K., G. Gkioxari, P. Dollar and R. Girshick. 2017. Mask R-CNN.
Pages 2980–2988 in
Proceedings of International Conference on
Computer Vision (ICCV)
, Venice, Italy.
Howard, A. G., M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T.
Weyand, M. Andreetto and H. Adam. 2017. MobileNets: efficient
convolutional neural networks for mobile vision applications.
CoRR
1704.
Hu, X., J. Shen, J. Shan, and L.J.I.G Pan. 2013. Local edge
distributions for detection of salient structure textures and
objects.
IEEE Geoscience and Remote Sensing Letters
10:466–
470.
Hu, Z., Q. Li, Q. Zhang and G. Wu. 2016. Representation of block-
based image features in a multi-scale framework for built-up area
detection.
Remote Sensing
8:155.
Huang, X. and L. Zhang. 2012. Morphological building/shadow
index for building extraction from high-resolution imagery over
urban areas.
IEEE Journal of Selected Topics in Applied Earth
Observations and Remote Sensing
5:161–172.
750
October 2019
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
