Graph based Semi-Supervised Learning with Class-Probability Distance for Hyperspectral Remote Sensing Image Classification

doi:10.13485/j.cnki.11-2089.2014.0182

Abstract

Abstract:

A class-probability distance based semi-supervised learning method is proposed for hyperspectral remote sensing image classification. In the method, Sparse Representation based Classification (SRC) is adopted for estimating the class-probability of unlabeled sample. Then a distance metric that describes the data similarity is developed based on the estimated class-probability. With this new distance metric, the distance between samples of different classes is enlarged effectively, and the neighbors of each sample can contain more samples belonging to the same class. Finally, this distance metric is applied to Linear Neighborhood Propagation and Label Propagation algorithms. Experimental results using Hyperion and AVIRIS hyperspectral remote sensing images show that the approach outperforms the existing semi-supervised learning methods in terms of classification accuracy.

Key words: hyperspectral remote sensing image classification, graph, semi-supervised learning, sparse representation, class-probability distance

CLC Number:

TP751

SHAO Yuanjie WU Guoping MA Li. Graph based Semi-Supervised Learning with Class-Probability Distance for Hyperspectral Remote Sensing Image Classification[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(11): 1182-1189.

References

[1] KIM W，CRAWFORD M M. Adaptive Classification for Hyperspectral Image Data Using Manifold Regularization Kernel Machines[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(11): 4110-4121.
[2] MILLER D J, UYAR H S. A Mixture of Experts Classifier with Learning Based on Both Labeled and Unlabelled Data[C]//Advances in Neural Information Processing Systems 9. Cambridge, MA:MIT Press. 1997. 571-577.
[3] NIGAM K, MCCALLUM A K, THRUM S, MITCHELL T. Text Classification from Labeled and Unlabeled Documents Using EM[J]. Machine Learning. 2000. 39 (2-3). 103-134.
[4] RATSABY J, VENKATESH S. Learning from a Mixture of Labeled and Unlabeled Examples with Parametric Side Information[C]//Proceedings of the Eighth Annual Conference on Computational Learning Theory. 1995. 412-417.
[5] CASTELLI V, COVER T. The Exponential Value of Labeled Samples[J]. Pattern Recognition Letters. 1995. 16 (1). 105-111.
[6] YAROWSKY D. Unsupervised Word Sense Disambiguation Rivaling Supervised Methods[C]// Proceedings of the 33rd annual meeting of the Association of Computational Linguistics. Cambridge, MA. 1995. I 89-196.
[7] RILOFF E, WIEBE J, WILSON T. Learning Subjective Nouns Using Extraction Pattern Bootstrapping[C]// Proceedings of the 7th Conference on Natural Language Learning. 2003.
[8] ROSENBERG C, HEBERT M, SCHNEIDERMAN H. Semi-supervised Self-training of Object Detection Models[C]//Proceedings.of the 7th IEEE Workshop on Applications of Computer Vision.2005.
[9] BLUM A, MITCHELL T. Combining Labeled and Lnlabeled Data with Co-training[C]//Proceedings of the 11thAnnual Conference on Learning Theory. Madison, WI. 1998. 92-100.
[10] CHAPELLE O, ZIEN A. Semi-supervised Classification by Low Density Separation[C]//Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics. 2005. 57-64.
[11] JOACHIMS T. Transductive Inference for Text Classification Using Support Vector Machines[C]//Proceedings of the 16th International Conference on Machine Learning. Bled,Slovenia. 1999. 200-209.
[12] ZHU X J, LAFFERTY J, GHAHRAMANI Z. Semi-supervised Learning using Gaussian Fields and Harmonic Functions[C]//Proceedings of the 20th International Conference on Machine Learning, Menlo Park, California: AAAI, 2003: 912-919.
[13] ZHOU D Y, BOUSQUET O, LAL T, et al. Learning with Local and Global Consistency[C]//Advances in Neural Information Processing Systems 16, Massachusetts, 2004, 16: 321-328.
[14] WANG F, ZHANG C S. Label Propagation through Linear Neighborhoods[J]. IEEE Transactions on Knowledge and Data Engineering, 2008, 20(1): 55-67.
[15] ZHAO Yinghai, CAI Junjie, WU Xiuqing, et al. Sparse Graph Based Transductive Multi-label Learning for Video Concept Detection[J]. Jounal of Pattern Recognition & Artificial Intelligence, 2011, 24(6): 825-831. (赵英海, 蔡俊杰, 吴秀清, 等. 基于稀疏化图结构的转导多标注视频概念检测算法[J]. 模式识别与人工智能, 2011, 24(6):825-831.)
[16] ROHBAN M H , RABIEE H R.Supervised Neighborhood Graph Construction for Semi-supervised Classification[J]. Pattern Recognition Letters, 2012, 45(4):1363-1372.
[17] KOBAYASHI T, WATANABE K , OTSU N. Logistic Label Propagation[J]. Pattern Recognition Letters, 2012, 33(5):580-588.
[18] ZHU X J, GOLDBERG A B. Introduction to Semi-supervised Learning [M]. USA:Morgan & Claypool, 2009.
[19] WRIGHT J, YANG A Y, GANESH A, et al. Robust Face Recognition via Sparse Representation[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2009, 312(2):210-227.
[20] RIDDER D D, KOUROPTEVA O, OKUN O, et al. Supervised Locally Linear Embedding[J]. Lecture Notes in Computer Science, 2003: 333-341.
[21] GAO Hengzhen. Research on Classification Techinique for Hyperspectral Remote Sensing Imagery[D].[Ph.D.dissertation],Nat-
ional University of Defense Technology, 2011.(高恒振. 高光谱遥感图像分类技术研究 [D].[博士论文]. 国防科学技术大学,2011.)