Unsupervised Band Selection by Integrating the Overall Accuracy and Redundancy

Band selection is of great significance to alleviate the curse of dimensionality for hyperspectral (HSI) image application. In this letter, we propose a novel unsupervised band selection method for HSI classification. This method integrates both the overall accuracy and redundancy into the band selection process by formulating an optimization model. In the optimization problem, an adaptive balance parameter is designed to trade off the overall accuracy and redundancy. Additionally, we adopt an unsupervised overall accuracy prediction method to obtain the overall accuracy; thus, no ground truth or training samples is required. Experimental results on the ROSIS and RetigaEx data sets show that our method outperforms four representative methods in terms of classification accuracy and redundancy.

[1]  Jun Zhou,et al.  A hypergraph based semi-supervised band selection method for hyperspectral image classification , 2013, 2013 IEEE International Conference on Image Processing.

[2]  Jianzhong Wang,et al.  Maximum weight and minimum redundancy: A novel framework for feature subset selection , 2013, Pattern Recognit..

[3]  Chein-I Chang,et al.  Constrained band selection for hyperspectral imagery , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Chein-I Chang,et al.  Hyperspectral Data Exploitation , 2007 .

[5]  Chein-I. Chang.  Hyperspectral Data Exploitation: Theory and Applications , 2007 .

[6]  Adolfo Martínez Usó,et al.  Clustering-based multispectral band selection using mutual information , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[7]  Robert I. Damper,et al.  A fast separability-based feature-selection method for high-dimensional remotely sensed image classification , 2008, Pattern Recognit..

[8]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[9]  Qian Du,et al.  Band selection and its impact on target detection and classification in hyperspectral image analysis , 2003, IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003.

[10]  Hairong Qi,et al.  Sparse representation based band selection for hyperspectral images , 2011, 2011 18th IEEE International Conference on Image Processing.

[11]  Qian Du,et al.  A joint band prioritization and band-decorrelation approach to band selection for hyperspectral image classification , 1999, IEEE Trans. Geosci. Remote. Sens..

[12]  Chenhong Sui,et al.  An Unsupervised Band Selection Method Based on Overall Accuracy Prediction , 2014, 2014 22nd International Conference on Pattern Recognition.

[13]  Lorenzo Bruzzone,et al.  A multiscale expectation-maximization semisupervised classifier suitable for badly posed image classification , 2006, IEEE Transactions on Image Processing.

[14]  Driss Aboutajdine,et al.  A Dashboard to Analysis and Synthesis of Dimensionality Reduction Methods in Remote Sensing , 2013, ArXiv.

[15]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[16]  B. Guo,et al.  Hyperspectral image fusion using spectrally weighted kernels , 2005, 2005 7th International Conference on Information Fusion.

[17]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[18]  Robert I. Damper,et al.  Band Selection for Hyperspectral Image Classification Using Mutual Information , 2006, IEEE Geoscience and Remote Sensing Letters.

[19]  Qian Du,et al.  An Efficient Method for Supervised Hyperspectral Band Selection , 2011, IEEE Geoscience and Remote Sensing Letters.

[20]  Anil K. Jain,et al.  Unsupervised Learning of Finite Mixture Models , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Shiming Xiang,et al.  Classification oriented semi-supervised band selection for hyperspectral images , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[22]  Jihao Yin,et al.  Optimal Band Selection for Hyperspectral Image Classification Based on Inter-Class Separability , 2010, 2010 Symposium on Photonics and Optoelectronics.