Hyperspectral band selection based on consistency-measure of neighborhood rough set theory

Band selection is a well-known approach for reducing dimensionality in hyperspectral imaging. In this paper, a band selection method based on consistency-measure of neighborhood rough set theory (CMNRS) was proposed to select informative bands from hyperspectral images. A decision-making information system was established by the reflection spectrum of soybeans' hyperspectral data between 400 nm and 1000 nm wavelengths. The neighborhood consistency-measure, which reflects not only the size of the decision positive region, but also the sample distribution in the boundary region, was used as the evaluation function of band significance. The optimal band subset was selected by a forward greedy search algorithm. A post-pruning strategy was employed to overcome the over-fitting problem and find the minimum subset. To assess the effectiveness of the proposed band selection technique, two classification models (extreme learning machine (ELM) and random forests (RF)) were built. The experimental results showed that the proposed algorithm can effectively select key bands and obtain satisfactory classification accuracy.

[1]  Keinosuke Fukunaga,et al.  A Branch and Bound Algorithm for Feature Subset Selection , 1977, IEEE Transactions on Computers.

[2]  Jihoon Yang,et al.  Feature Subset Selection Using a Genetic Algorithm , 1998, IEEE Intell. Syst..

[3]  Huan Liu,et al.  A Monotonic Measure for Optimal Feature Selection , 1998, ECML.

[4]  Philip S. Yu.  IEEE Transactions on Knowledge and Data Engineering: EIC Editorial , 2001 .

[5]  Chong-Ho Choi,et al.  Input feature selection for classification problems , 2002, IEEE Trans. Neural Networks.

[6]  Zdzislaw Pawlak,et al.  Rough sets and intelligent data analysis , 2002, Inf. Sci..

[7]  Huan Liu,et al.  Consistency-based search in feature selection , 2003, Artif. Intell..

[8]  Andrzej Skowron,et al.  Rough set methods in feature selection and recognition , 2003, Pattern Recognit. Lett..

[9]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

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

[11]  Huan Liu,et al.  Toward integrating feature selection algorithms for classification and clustering , 2005, IEEE Transactions on Knowledge and Data Engineering.

[12]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Adrian J. Brown.  Spectral curve fitting for automatic hyperspectral data analysis , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[14]  Qinghua Hu,et al.  Information-preserving hybrid data reduction based on fuzzy-rough techniques , 2006, Pattern Recognit. Lett..

[15]  Yee Leung,et al.  A rough set approach to the discovery of classification rules in spatial data , 2007, Int. J. Geogr. Inf. Sci..

[16]  Zexuan Zhu,et al.  Wrapper–Filter Feature Selection Algorithm Using a Memetic Framework , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[17]  Qinghua Hu,et al.  Hybrid attribute reduction based on a novel fuzzy-rough model and information granulation , 2007, Pattern Recognit..

[18]  Xiangyang Wang,et al.  Feature selection based on rough sets and particle swarm optimization , 2007, Pattern Recognit. Lett..

[19]  Qinghua Hu,et al.  Consistency Based Attribute Reduction , 2007, PAKDD.

[20]  Xianfeng Fan,et al.  Machine fault feature extraction based on intrinsic mode functions , 2008 .

[21]  Qinghua Hu,et al.  Neighborhood rough set based heterogeneous feature subset selection , 2008, Inf. Sci..

[22]  Seoung Bum Kim,et al.  Genetic algorithm-based feature selection in high-resolution NMR spectra , 2008, Expert Syst. Appl..

[23]  S. Dunagan,et al.  The MARTE VNIR imaging spectrometer experiment: design and analysis. , 2008, Astrobiology.

[24]  Jacek M. Zurada,et al.  Normalized Mutual Information Feature Selection , 2009, IEEE Transactions on Neural Networks.

[25]  Yun Li,et al.  Feature selection based on loss-margin of nearest neighbor classification , 2009, Pattern Recognit..

[26]  Qinghua Hu,et al.  Feature evaluation and selection based on neighborhood soft margin , 2010, Neurocomputing.

[27]  Witold Pedrycz,et al.  Selecting Discrete and Continuous Features Based on Neighborhood Decision Error Minimization , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[28]  Xin Pan,et al.  A variable precision rough set approach to the remote sensing land use/cover classification , 2010, Comput. Geosci..

[29]  Yi Lin,et al.  Optimizing model for land use/land cover retrieval from remote sensing imagery based on variable precision rough sets , 2011 .

[30]  Nan Liu,et al.  Voting based extreme learning machine , 2012, Inf. Sci..

[31]  Zhi-gang Su,et al.  Minimizing neighborhood evidential decision error for feature evaluation and selection based on evidence theory , 2012, Expert Syst. Appl..

[32]  Qinghua Hu,et al.  Mechanical fault diagnosis based on redundant second generation wavelet packet transform, neighborhood rough set and support vector machine , 2012 .

[33]  Ouen Pinngern,et al.  Feature subset selection wrapper based on mutual information and rough sets , 2012, Expert Syst. Appl..

[34]  Qingxin Zhu,et al.  Quantitative analysis for covering-based rough sets through the upper approximation number , 2013, Inf. Sci..

[35]  Tan Kezhu,et al.  Detection of Isoflavones Content in Soybean Based on Hyperspectral Imaging Technology , 2014 .

[36]  Jiye Liang,et al.  International Journal of Approximate Reasoning Multigranulation Decision-theoretic Rough Sets , 2022 .

[37]  Yao Liu,et al.  Hyperspectral band selection based on a variable precision neighborhood rough set. , 2016, Applied optics.

[38]  Yu Peng,et al.  MWPCA-ICURD: density-based clustering method discovering specific shape original features , 2016, Neural Computing and Applications.