Advanced search
1 file | 17.47 MB Add to list

Joint classification of hyperspectral and LiDAR data using hierarchical random walk and deep CNN architecture

Author
Organization
Abstract
Earth observation using multisensor data is drawing increasing attention. Fusing remotely sensed hyperspectral imagery and light detection and ranging (LiDAR) data helps to increase application performance. In this article, joint classification of hyperspectral imagery and LiDAR data is investigated using an effective hierarchical random walk network (HRWN). In the proposed HRWN, a dual-tunnel convolutional neural network (CNN) architecture is first developed to capture spectral and spatial features. A pixelwise affinity branch is proposed to capture the relationships between classes with different elevation information from LiDAR data and confirm the spatial contrast of classification. Then in the designed hierarchical random walk layer, the predicted distribution of dual-tunnel CNN serves as global prior while pixelwise affinity reflects the local similarity of pixel pairs, which enforce spatial consistency in the deeper layers of networks. Finally, a classification map is obtained by calculating the probability distribution. Experimental results validated with three real multisensor remote sensing data demonstrate that the proposed HRWN significantly outperforms other state-of-the-art methods. For example, the two branches CNN classifier achieves an accuracy of 88.91 on the University of Houston campus data set, while the proposed HRWN classifier obtains an accuracy of 93.61, resulting in an improvement of approximately 5.
Keywords
Electrical and Electronic Engineering, General Earth and Planetary Sciences, Feature extraction, Laser radar, Hyperspectral imaging, Convolution, Probability distribution, Convolutional neural network (CNN), hyperspectral image (HSI), multisensor data fusion, hierarchical random walk, IMAGE CLASSIFICATION, EXTINCTION PROFILES, WAVE-FORM, FUSION

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 17.47 MB

Citation

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

MLA
Zhao, Xudong, et al. “Joint Classification of Hyperspectral and LiDAR Data Using Hierarchical Random Walk and Deep CNN Architecture.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 58, no. 10, 2020, pp. 7355–70, doi:10.1109/tgrs.2020.2982064.
APA
Zhao, X., Tao, R., Li, W., Li, H.-C., Du, Q., Liao, W., & Philips, W. (2020). Joint classification of hyperspectral and LiDAR data using hierarchical random walk and deep CNN architecture. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 58(10), 7355–7370. https://doi.org/10.1109/tgrs.2020.2982064
Chicago author-date
Zhao, Xudong, Ran Tao, Wei Li, Heng-Chao Li, Qian Du, Wenzhi Liao, and Wilfried Philips. 2020. “Joint Classification of Hyperspectral and LiDAR Data Using Hierarchical Random Walk and Deep CNN Architecture.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 58 (10): 7355–70. https://doi.org/10.1109/tgrs.2020.2982064.
Chicago author-date (all authors)
Zhao, Xudong, Ran Tao, Wei Li, Heng-Chao Li, Qian Du, Wenzhi Liao, and Wilfried Philips. 2020. “Joint Classification of Hyperspectral and LiDAR Data Using Hierarchical Random Walk and Deep CNN Architecture.” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 58 (10): 7355–7370. doi:10.1109/tgrs.2020.2982064.
Vancouver
1.
Zhao X, Tao R, Li W, Li H-C, Du Q, Liao W, et al. Joint classification of hyperspectral and LiDAR data using hierarchical random walk and deep CNN architecture. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING. 2020;58(10):7355–70.
IEEE
[1]
X. Zhao et al., “Joint classification of hyperspectral and LiDAR data using hierarchical random walk and deep CNN architecture,” IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 58, no. 10, pp. 7355–7370, 2020.
@article{8658091,
  abstract     = {{Earth observation using multisensor data is drawing increasing attention. Fusing remotely sensed hyperspectral imagery and light detection and ranging (LiDAR) data helps to increase application performance. In this article, joint classification of hyperspectral imagery and LiDAR data is investigated using an effective hierarchical random walk network (HRWN). In the proposed HRWN, a dual-tunnel convolutional neural network (CNN) architecture is first developed to capture spectral and spatial features. A pixelwise affinity branch is proposed to capture the relationships between classes with different elevation information from LiDAR data and confirm the spatial contrast of classification. Then in the designed hierarchical random walk layer, the predicted distribution of dual-tunnel CNN serves as global prior while pixelwise affinity reflects the local similarity of pixel pairs, which enforce spatial consistency in the deeper layers of networks. Finally, a classification map is obtained by calculating the probability distribution. Experimental results validated with three real multisensor remote sensing data demonstrate that the proposed HRWN significantly outperforms other state-of-the-art methods. For example, the two branches CNN classifier achieves an accuracy of 88.91 on the University of Houston campus data set, while the proposed HRWN classifier obtains an accuracy of 93.61, resulting in an improvement of approximately 5.}},
  author       = {{Zhao, Xudong and Tao, Ran and Li, Wei and Li, Heng-Chao and Du, Qian and Liao, Wenzhi and Philips, Wilfried}},
  issn         = {{0196-2892}},
  journal      = {{IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING}},
  keywords     = {{Electrical and Electronic Engineering,General Earth and Planetary Sciences,Feature extraction,Laser radar,Hyperspectral imaging,Convolution,Probability distribution,Convolutional neural network (CNN),hyperspectral image (HSI),multisensor data fusion,hierarchical random walk,IMAGE CLASSIFICATION,EXTINCTION PROFILES,WAVE-FORM,FUSION}},
  language     = {{eng}},
  number       = {{10}},
  pages        = {{7355--7370}},
  title        = {{Joint classification of hyperspectral and LiDAR data using hierarchical random walk and deep CNN architecture}},
  url          = {{http://dx.doi.org/10.1109/tgrs.2020.2982064}},
  volume       = {{58}},
  year         = {{2020}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: