Automatic design of convolutional neural network for hyperspectral image classification

Yushi Chen*, Kaiqiang Zhu, Lin Zhu, Xin He, Pedram Ghamisi, Jon Atli Benediktsson

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

41 Citations (Scopus)

Abstract

Hyperspectral image (HSI) classification is a core task in the remote sensing community, and recently, deep learning-based methods have shown their capability of accurate classification of HSIs. Among the deep learning-based methods, deep convolutional neural networks (CNNs) have been widely used for the HSI classification. In order to obtain a good classification performance, substantial efforts are required to design a proper deep learning architecture. Furthermore, the manually designed architecture may not fit a specific data set very well. In this paper, the idea of automatic CNN for the HSI classification is proposed for the first time. First, a number of operations, including convolution, pooling, identity, and batch normalization, are selected. Then, a gradient descent-based search algorithm is used to effectively find the optimal deep architecture that is evaluated on the validation data set. After that, the best CNN architecture is selected as the model for the HSI classification. Specifically, the automatic 1-D Auto-CNN and 3-D Auto-CNN are used as spectral and spectral-spatial HSI classifiers, respectively. Furthermore, the cutout is introduced as a regularization technique for the HSI spectral-spatial classification to further improve the classification accuracy. The experiments on four widely used hyperspectral data sets (i.e., Salinas, Pavia University, Kennedy Space Center, and Indiana Pines) show that the automatically designed data-dependent CNNs obtain competitive classification accuracy compared with the state-of-the-art methods. In addition, the automatic design of the deep learning architecture opens a new window for future research, showing the huge potential of using neural architectures' optimization capabilities for the accurate HSI classification.

Original languageEnglish
Article number8703410
Pages (from-to)7048-7066
Number of pages19
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume57
Issue number9
DOIs
Publication statusPublished - Sept 2019

Bibliographical note

Funding Information:
Manuscript received January 1, 2019; revised March 17, 2019; accepted April 7, 2019. Date of publication April 30, 2019; date of current version August 27, 2019. This work was supported by the Natural Science Foundation of China under Grant 61771171. (Corresponding author: Yushi Chen.) Y. Chen, K. Zhu, L. Zhu, and X. He are with the School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China (e-mail: chenyushi@hit.edu.cn; zhukq1995@163.com; 17s105139@stu.hit.edu.cn; 1091636421@qq.com).

Publisher Copyright:
© 1980-2012 IEEE.

Other keywords

  • Convolutional neural network (CNN)
  • deep learning
  • hyperspectral image (HSI) classification
  • neural architecture search (NAS)

Fingerprint

Dive into the research topics of 'Automatic design of convolutional neural network for hyperspectral image classification'. Together they form a unique fingerprint.

Cite this