A Spatial Hypergraph Based Semi-Supervised Band Selection Method for Hyperspectral Imagery Semantic Interpretation

Hyperspectral imagery (HSI) typically provides a wealth of information captured in a wide range of the electromagnetic spectrum for each pixel in the image. Hence, a pixel in HSI is a high-dimensional vector of intensities with a large spectral range and a high spectral resolution. Therefore, the semantic interpretation is a challenging task of HSI analysis. We focused in this paper on object classification as HSI semantic interpretation. However, HSI classification still faces some issues, among which are the following: The spatial variability of spectral signatures, the high number of spectral bands, and the high cost of true sample labeling. Therefore, the high number of spectral bands and the low number of training samples pose the problem of the curse of dimensionality. In order to resolve this problem, we propose to introduce the process of dimensionality reduction trying to improve the classification of HSI. The presented approach is a semi-supervised band selection method based on spatial hypergraph embedding model to represent higher order relationships with different weights of the spatial neighbors corresponding to the centroid of pixel. This semi-supervised band selection has been developed to select useful bands for object classification. The presented approach is evaluated on AVIRIS and ROSIS HSIs and compared to other dimensionality reduction methods. The experimental results demonstrate the efficacy of our approach compared to many existing dimensionality reduction methods for HSI classification.

[1]  Steffen Staab,et al.  Comparative evaluation of spatial context techniques for semantic image analysis , 2009, 2009 10th Workshop on Image Analysis for Multimedia Interactive Services.

[2]  James E. Fowler,et al.  Locality-Preserving Dimensionality Reduction and Classification for Hyperspectral Image Analysis , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[3]  Qingshan Liu,et al.  Image retrieval via probabilistic hypergraph ranking , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Feiping Nie,et al.  A general graph-based semi-supervised learning with novel class discovery , 2010, Neural Computing and Applications.

[5]  Masashi Sugiyama,et al.  Dimensionality Reduction of Multimodal Labeled Data by Local Fisher Discriminant Analysis , 2007, J. Mach. Learn. Res..

[6]  Mihai Datcu,et al.  Semantic interpretation of multi-level change detection in multi-temporal satellite images , 2015, 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[7]  Filiberto Pla,et al.  Supervised feature selection by clustering using conditional mutual information-based distances , 2010, Pattern Recognit..

[8]  Wu Jian.  Unsupervised Intrusion Feature Selection based on Genetic Algorithm and FCM , 2012 .

[9]  Fang Liu,et al.  Unsupervised feature selection based on maximum information and minimum redundancy for hyperspectral images , 2016, Pattern Recognit..

[10]  Stephen Lin,et al.  Graph Embedding and Extensions: A General Framework for Dimensionality Reduction , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[12]  Yuan Yan Tang,et al.  Learning With Hypergraph for Hyperspectral Image Feature Extraction , 2015, IEEE Geoscience and Remote Sensing Letters.

[13]  Bor-Chen Kuo,et al.  Kernel Nonparametric Weighted Feature Extraction for Hyperspectral Image Classification , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[14]  Jiamin Liu,et al.  Enhanced semi-supervised local Fisher discriminant analysis for face recognition , 2012, Future Gener. Comput. Syst..

[15]  Hamid R. Rabiee,et al.  Spatial-Aware Dictionary Learning for Hyperspectral Image Classification , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[16]  Xuelong Li,et al.  Spectral-Spatial Constraint Hyperspectral Image Classification , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[17]  Henri Luchian,et al.  A unifying criterion for unsupervised clustering and feature selection , 2011, Pattern Recognit..

[18]  M. Thonnat,et al.  Symbol Grounding for Semantic Image Interpretation: From Image Data to Semantics , 2005, Tenth IEEE International Conference on Computer Vision Workshops (ICCVW'05).

[19]  Seoung Bum Kim,et al.  Unsupervised feature selection using weighted principal components , 2011, Expert Syst. Appl..

[20]  Fethi Ben Ouezdou,et al.  Stability of Dimensionality Reduction Methods Applied on Artificial Hyperspectral Images , 2012, ICCVG.

[21]  Chun Chen,et al.  A Unified Feature and Instance Selection Framework Using Optimum Experimental Design , 2012, IEEE Transactions on Image Processing.

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

[23]  Mile Pavlic,et al.  Data preparation for semantic image interpretation , 2010, Proceedings of the ITI 2010, 32nd International Conference on Information Technology Interfaces.

[24]  P. Deepa,et al.  Feature extraction of hyperspectral image using principal component analysis and folded-principal component analysis , 2015, 2015 2nd International Conference on Electronics and Communication Systems (ICECS).

[25]  Marina Ivasic-Kos,et al.  The analysis and overview of semantic image interpretation , 2009, Proceedings of the ITI 2009 31st International Conference on Information Technology Interfaces.

[26]  Hongyi Li,et al.  ISOMAP-based subspace analysis for the classification of hyperspectral data , 2013, 2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS.

[27]  Daoqiang Zhang,et al.  Semisupervised Dimensionality Reduction With Pairwise Constraints for Hyperspectral Image Classification , 2011, IEEE Geoscience and Remote Sensing Letters.