The classification algorithms of polarimetric synthetic aperture radar (PolSAR) imagesare generally composed of the feature extractors that transform the raw data into discriminative representations, followed by trainable classifiers. Traditional approaches always suffer from the hand-designed features and misclassification of boundary pixels. Following the great success of convolutional neural network (CNN), a novel data-driven classification framework based on the fusion of CNN and superpixel algorithm is presented in this paper. First, the region-based complex-valued network utilizes both the intensity and phase information to predict the label of each pixel and constructs the label map based on spatial relations. Second, superpixel generating algorithm is adopted to produce the superpixel representation of the Pauli decomposition image, and the contour information which reflects the boundary of each category is preserved. Finally, the original label map and contour information are fused to make the decision of each pixel, outputting the final label map. Experimental results on public datasets illustrate that the proposed method can automatically learn the intrinsic features from the PolSAR image for classification purpose. Besides, the fusion of the superpixel features can effectively correct the misclassification of the boundary and singular pixels, thus achieving superior performance.
2. Scheuchl, B., D. Flett, R. Caves, and I. Cumming, "Potential of RADARSAT-2 data for operational sea ice monitoring," Canadian Journal of Remote Sensing, Vol. 30, No. 3, 448-461, 2004.
3. Wang, L., et al., "Sea ice concentration estimation during melt from dualpol SAR scenes using deep convolutional neural networks: A case study," IEEE Transactions on Geoscience & Remote Sensing, Vol. 54, No. 8, 4524-4533, 2016.
4. Freeman, A., et al., "On the use of multi-frequency and polarimetric radar backscatter features for classification of agricultural crops," International Journal of Remote Sensing, Vol. 15, No. 9, 14, 1994.
5. Lee, J. S. and M. R. Grunes, "Classification of multi-look polarimetric SAR data based on complex Wishart distribution," National Telesystems Conference, IEEE, 1992.
6. Gao, W., J. Yang, and W. Ma, "Land cover classification for polarimetric SAR images based on mixture models," Remote Sensing, Vol. 6, No. 5, 3770-3790, 2014.
7. Rignot, E. and R. Chellappa, "Segmentation of polarimetric synthetic aperture radar data," IEEE Transactions on Image Processing, Vol. 1, No. 3, 281-300, 1992.
8. Lee, J. S., et al., "K-distribution for multi-look processed polarimetric SAR imagery," International Geoscience & Remote Sensing Symposium, IEEE, 1994.
9. Freitas, C. C., A. C. Frery, and A. H. Correia, "The polarimetric distribution for sar data analysis," Environmetrics, Vol. 16, No. 1, 13-31, 2010.
10. Chen, Q., et al., "Unsupervised land cover/land use classification using PolSAR imagery based on scattering similarity," IEEE Transactions on Geoscience & Remote Sensing, Vol. 51, No. 3, 1817-1825, 2013.
11. Wang, Y., C. Han, and F. Tupin, "PolSAR data segmentation by combining tensor space cluster analysis and Markovian framework," IEEE Geoscience & Remote Sensing Letters, Vol. 7, No. 1, 210-214, 2010.
12. Shang, F. and A. Hirose, "Quaternion neural-network-based PolSAR land classification in Poincaresphere-parameter space," IEEE Transactions on Geoscience & Remote Sensing, Vol. 52, No. 9, 5693-5703, 2014.
13. Yu, P., A. K. Qin, and D. A. Clausi, "Unsupervised polarimetric SAR image segmentation and classification using region growing with edge penalty," IEEE Transactions on Geoscience & Remote Sensing, Vol. 50, No. 4, 1302-1317, 2012.
14. Cao, F., W. Hong, Y. Wu, and E. Pottier, "An unsupervised segmentation with an adaptive number of clusters using the SPAN/H/α/A space and the complex Wishart clustering for fully polarimetric SAR data analysis," IEEE Transactions on Geoscience & Remote Sensing, Vol. 45, No. 11, 3454-3467, 2007.
15. Wu, Y., K. Ji, W. Yu, and Y. Su, "Region-based classification of polarimetric SAR images using Wishart MRF," IEEE Geoscience & Remote Sensing Letters, Vol. 5, No. 4, 668-672, 2008.
16. Hou, B., H. Kou, and L. Jiao, "Classification of polarimetric SAR images using multilayer autoencoders and superpixels," IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, Vol. 9, No. 7, 3072-3081, 2017.
17. Krizhevsky, A., I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," Proc. Adv. Neural Inf. Process. Syst., 1097-1105, 2012.
18. He, K., X. Zhang, S. Ren, and J. Sun, Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification, 2015.
19. Li, Q., et al., "Medical image classification with convolutional neural network," International Conference on Control Automation Robotics and Vision, 844-848, IEEE, 2014.
20. Chen, S., H. Wang, F. Xu, and Y.-Q. Jin, "Target classification using the deep convolutional networks for SAR images," IEEE Transactions on Geoscience & Remote Sensing, Vol. 54, No. 8, 4806-4817, 2016.
21. Zhou, Y., H. Wang, F. Xu, and Y.-Q. Jin, "Polarimetric SAR image classification using deep convolutional neural networks," IEEE Geoscience & Remote Sensing Letters, Vol. 13, No. 12, 1935-1939, 2016.
22. Hirose, A., Complex-Valued Neural Networks: Advances and Applications, Wiley, 2013.
23. Zhang, Z., et al., "Complex-valued convolutional neural network and its application in polarimetric SAR image classification," IEEE Transactions on Geoscience & Remote Sensing, Vol. 99, 1-12, 2017.
24. Zeiler, M. D., et al., "Adaptive deconvolutional networks for mid and high level featurelearning," International Conference on Computer Vision, 2018-2025, 2011.
25. Ren, X. and J. Malik, "Learning a classification model for segmentation," International Conference on Computer Vision, Vol. 1, 10-17, 2003.
26. Cloude, S. R. and E. Pottier, "A review of target decomposition theorems in radar polarimetry," IEEE Transactions on Geoscience & Remote Sensing, Vol. 34, No. 2, 498-518, 1996.
27. Zhu, B., et al., "Image reconstruction by domain-transform manifold learning," Nature, 2017.
28. Lecun, Y., L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, Vol. 86, No. 11, 2278-2324, 1998.
29. Zhang, Y., W. Miao, Z. Lin, H. Gao, and S. Shi, "Millimeter-wave InSAR image reconstruction approach by total variation regularized matrix completion," Remote Sens., Vol. 10, 1053, 2018.