Hyperspectral Image Clustering based on Variational Expectation Maximization
ID:27 View Protection:ATTENDEE Updated Time:2020-08-05 10:16:59 Hits:449 Oral Presentation

Start Time:2020-06-09 14:45(Asia/Shanghai)


Session:R Regular Session » R06Machine Learning-Based Multi-channel Processing

Video No Permission

Tips: Only the registered participant can access the file. Please sign in first.

Hyperspectral image clustering is an important and challenging problem, which aims to group image pixels according to the land cover information extracted from the spectrum. The spectrum observed at adjacent pixels are often highly-correlated, and leveraging such spatial correlation can greatly improve the clustering accuracy. Markov Random Field (MRF) is a powerful model to characterize such correlation. However, in this model the spatial parameter \beta often needs to be manually tuned, which brings difficulty in finding an optimal value. In this paper, we propose a novel hyperspectral clustering algorithm, which is able to learn parameter \beta from the data and thus achieves better performance. Specifically, we model the spectral information with Gaussian mixture model, and use variational expectation maximization method to complete the parameter estimation and clustering task. Experiments on both synthetic and real-world data sets verify the effectiveness of the proposed algorithm.
Hyperspectral Image clustering; undirected graphical model; variational expectation maximization,; spatial parameter
Yuchen Jiao
Tsinghua University, China

Submission Author
Yuchen Jiao Tsinghua University, China
Yirong Ma 69010 Unit PLA, China
Yuantao Gu Tsinghua University, China
Submit Comment
Verify Code Change Another
All Comments
Important Date
  • Conference Date

    Jun 08



    Jun 11


  • Jan 12 2020

    Draft paper submission deadline

  • Apr 15 2020

    Early Bird Registration

Sponsored By
IEEE Signal Processing Society
Organized By
Zhejiang University
Contact Information