Advanced search
1 file | 2.67 MB

High-resolution aerial imagery semantic labeling with dense pyramid network

(2018) SENSORS. 18(11). p.1-15
Author
Organization
Abstract
Semantic segmentation of high-resolution aerial images is of great importance in certain fields, but the increasing spatial resolution brings large intra-class variance and small inter-class differences that can lead to classification ambiguities. Based on high-level contextual features, the deep convolutional neural network (DCNN) is an effective method to deal with semantic segmentation of high-resolution aerial imagery. In this work, a novel dense pyramid network (DPN) is proposed for semantic segmentation. The network starts with group convolutions to deal with multi-sensor data in channel wise to extract feature maps of each channel separately; by doing so, more information from each channel can be preserved. This process is followed by the channel shuffle operation to enhance the representation ability of the network. Then, four densely connected convolutional blocks are utilized to both extract and take full advantage of features. The pyramid pooling module combined with two convolutional layers are set to fuse multi-resolution and multi-sensor features through an effective global scenery prior manner, producing the probability graph for each class. Moreover, the median frequency balanced focal loss is proposed to replace the standard cross entropy loss in the training phase to deal with the class imbalance problem. We evaluate the dense pyramid network on the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam 2D semantic labeling dataset, and the results demonstrate that the proposed framework exhibits better performances, compared to the state of the art baseline.

Downloads

  • sensors-18-03774.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 2.67 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Pan, Xuran , Lianru Gao, Bing Zhang, Fan Yang, and Wenzhi Liao. 2018. “High-resolution Aerial Imagery Semantic Labeling with Dense Pyramid Network.” Sensors 18 (11): 1–15.
APA
Pan, X., Gao, L., Zhang, B., Yang, F., & Liao, W. (2018). High-resolution aerial imagery semantic labeling with dense pyramid network. SENSORS, 18(11), 1–15.
Vancouver
1.
Pan X, Gao L, Zhang B, Yang F, Liao W. High-resolution aerial imagery semantic labeling with dense pyramid network. SENSORS. MDPI; 2018;18(11):1–15.
MLA
Pan, Xuran et al. “High-resolution Aerial Imagery Semantic Labeling with Dense Pyramid Network.” SENSORS 18.11 (2018): 1–15. Print.
@article{8604468,
  abstract     = {Semantic segmentation of high-resolution aerial images is of great importance in certain
fields, but the increasing spatial resolution brings large intra-class variance and small inter-class
differences that can lead to classification ambiguities. Based on high-level contextual features, the deep
convolutional neural network (DCNN) is an effective method to deal with semantic segmentation of
high-resolution aerial imagery. In this work, a novel dense pyramid network (DPN) is proposed for
semantic segmentation. The network starts with group convolutions to deal with multi-sensor data in
channel wise to extract feature maps of each channel separately; by doing so, more information from
each channel can be preserved. This process is followed by the channel shuffle operation to enhance
the representation ability of the network. Then, four densely connected convolutional blocks are
utilized to both extract and take full advantage of features. The pyramid pooling module combined
with two convolutional layers are set to fuse multi-resolution and multi-sensor features through an
effective global scenery prior manner, producing the probability graph for each class. Moreover,
the median frequency balanced focal loss is proposed to replace the standard cross entropy loss in the
training phase to deal with the class imbalance problem. We evaluate the dense pyramid network on
the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam
2D semantic labeling dataset, and the results demonstrate that the proposed framework exhibits
better performances, compared to the state of the art baseline.},
  articleno    = {3774},
  author       = {Pan, Xuran  and Gao, Lianru and Zhang, Bing  and Yang, Fan  and Liao, Wenzhi},
  issn         = {1424-8220 },
  journal      = {SENSORS},
  language     = {eng},
  number       = {11},
  pages        = {3774:1--3774:15},
  title        = {High-resolution aerial imagery semantic labeling with dense pyramid network},
  url          = {http://dx.doi.org/10.3390/s18113774},
  volume       = {18},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: