Combined dictionary learning based sparse representation for PolSAR image classification

被引：0

作者：

Liu L. ^{[1
]}

Liu S. ^{[2
]}

Jiao L. ^{[2
]}

Jin S. ^{[2
]}

机构：

[1] School of Computer Science and Engineering, Xi'an University of Technology, Xi'an

[2] Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, International Research Center for Intelligent Perception and Computation, Xidian University, Xi'an

来源：

Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition) | 2016年 / 44卷 / 02期

关键词：

Affinity propagation clustering; Dictionary learning; Linear support vector machine; Manifold distance; PolSAR image classification; Sparse representation;

D O I：

10.13245/j.hust.160217

中图分类号：

学科分类号：

摘要：

Traditional dictionary learning (DL) algorithms only consider the global sparsity of data, yet ignore the spatial structure of data. Moreover, its high computational complexity leads to the difficulty of dealing with large-scale image data. Considering the information of PolSAR image in the spatial-polarimetric domain, a novel combined DL based sparse representation (SR) classification method (CDL-SRC) was proposed for PolSAR image classification in this paper. First, the spatial-polarimetric manifold based fast affinity propagation (AP) clustering was employed to learn an over-complete dictionary. Then locality-constrained linear coding method was adopted to extract the spatial and polarimetric features of PolSAR respectively. Finally, the PolSAR image was classified by the linear support vector machine (SVM). Compared with traditional methods, experimental results demonstrate that the proposed method can improve the classification accuracy, which has the advantages of strong adaptability, efficient convergence rate and low computational complexity. © 2016, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.

引用

页码：81 / 85

页数：4

共 10 条

[1]

Tao M.L., Zhou F., Liu Y., Et al., Tensorial independent component analysis-based feature extraction for polarimetric SAR data classification, IEEE Trans on Geosci Remote Sens, 53, 5, pp. 2481-2495, (2015)

[2]

Yang M., Zhang L., Feng X., Et al., Sparse representation based fisher discrimination dictionary learning for image classification, International Journal of Computer Vision, 109, 3, pp. 209-232, (2014)

[3]

He C., Liu M., Liao Z., Et al., A learning-based target decomposition method using Kernel KSVD for polarimetric SAR image classification, EURASIP Journal on Advances in Signal Processing, 1, pp. 1-9, (2012)

[4]

Cao W., Chen J., Richard C., Et al., Online dictionary learning for kernel LMS, IEEE Trans on Signal Processing, 62, 11, pp. 2765-2777, (2014)

[5]

Chen J., Li Q., Peng Q., Et al., CSIFT based locality-constrained linear coding for image classification, Pattern Analysis and Applications, 18, 2, pp. 441-450, (2015)

[6]

Ophir B., Lustig M., Elad M., Multi-scale dictionary learning using wavelets, IEEE Journal of Selected Topics in Signal Processing, 5, 5, pp. 1014-1024, (2011)

[7]

He C., Li S., Liao Z., Et al., Texture classification of PolSAR data based on sparse coding of wavelet polarization textons, IEEE Trans on Geosci Remote Sens, 51, 8, pp. 4576-4590, (2013)

[8]

Shang F., Jiao L., Shi J., Et al., Fast affinity propagation clustering: a multilevel approach, Pattern Recognition, 45, pp. 474-486, (2012)

[9]

Chang C.C., Lin C.J., LIBSVM: A library for support vector machines, ACM Transactions on Intelligent Systems and Technology (TIST), 2, 3, (2011)

[10]

Fukuda S., Hirosawa H., Support vector machine classification of land cover: application to polarimetric SAR data, Geoscience and Remote Sensing Symposium, pp. 187-189, (2001)

← 1 →