Sea ice change detection from synthetic aperture radar images

based on distinctive analysis and LSTM

Submitted to Remote Sensing Letters


Feng Gao, Xiao Wang, Junyu Dong, Shengke Wang

Last modified: 2018/1/15

Abstract: Polar sea ice change detection is important for navigation safety and has drawn great attentions in recent years. In this letter, distinctive analysis and long shortterm memory (LSTM) are integrated to solve problem of the sea ice change detection for synthetic aperture radar (SAR) images. Firstly, the distinctive analysis algorithm is utilized to refined the difference image generated by the log-ratio operator. Then, from the refined difference image, reliable samples are selected to train the LSTM network. The LSTM network is driven to learn the change information, and a powerful model is built to distinguish changed pixels from noisy background polluted by the speckle noise. Finally, pixels from the original SAR images are classified into changed or unchanged class by the LSTM network, and the final change map can be obtained. Experimental results on two real SAR datasets demonstrate the effectiveness and superiority of the proposed method over three closely related methods.

[ MATLAB code ]

The demo has not been well organized. Please contact me if you meet any problems.


Links to most related works

  1. Achanta et al., 2012. "SLIC superpixels compared to state-of-the-art superpixel methods." IEEE Transactions on Pattern Analysis and Machine Intelligence 34 (11): 2274-2282.
  2. Celik, T., 2009. "Unsupervised change detection in satellite images using principal component analysis and k-means clustering." IEEE Geoscience and Remote Sensing Letters 6 (4): 772-776.
  3. Gao, F., Dong, J., Li, B., Xu, Q., Xie, C., 2016. "Change detection from synthetic aperture radar images based on neighborhood-based ratio and extreme learning machine." Journal of Applied Remote Sensing 10 (4): 046019.
  4. Gao, F., Dong, J., Li, B., Xu, Q., 2016. "Automatic change detection in synthetic aperture radar images based on PCANet." IEEE Geoscience and Remote Sensing Letters 13 (12): 1792-1796.
  5. Gong, M., Jia, M., Su, L., Wang, S., Jian, L., 2014. "Detecting changes of the Yellow River Estuary via SAR images based on local fit-search model and kernel-induced Graph Cuts." International Journal of Remote Sensing 35 (11): 4009-4030.



>>   return to Feng Gao's homepage 



Last modified: 2018/1/15.