Segmentation of hyperspectral images using local covariance matrices

In this work, basically, the local covariance matrices are used for the purpose of unsupervised segmentation of the hyperspectral images and the effect on the segmentation accuracy is also observed. The acquisition of the hyperspectral images with label (or groundtruth) information is very expensive and time consuming process. For this reason, realizing segmentation without label information brings important advantage in the analysis of the hyperspectral images. Proposed local covariance matrices represent a combined approach for using both spatial and spectral information together which is very important in hyperspectral image processing area. In the simulations, information divergence band selection method for reducing computational complexity and the positive effects of the proposed approach were proven with the experiments.

[1]  Jon Atli Benediktsson,et al.  Segmentation and Classification of Hyperspectral Images Using Minimum Spanning Forest Grown From Automatically Selected Markers , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[2]  Marco Diani,et al.  An unsupervised algorithm for hyperspectral image segmentation based on the Gaussian mixture model , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[3]  Ujjwal Maulik,et al.  Multiobjective Genetic Clustering for Pixel Classification in Remote Sensing Imagery , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Luis O. Jimenez-Rodriguez,et al.  Unsupervised Linear Feature-Extraction Methods and Their Effects in the Classification of High-Dimensional Data , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Mohamad M. Awad,et al.  Multicomponent Image Segmentation Using a Genetic Algorithm and Artificial Neural Network , 2007, IEEE Geoscience and Remote Sensing Letters.

[6]  F. Meer.  The effectiveness of spectral similarity measures for the analysis of hyperspectral imagery , 2006 .

[7]  David A. Landgrebe,et al.  Signal Theory Methods in Multispectral Remote Sensing , 2003 .

[8]  Russell M. Mersereau,et al.  Robust automatic clustering of hyperspectral imagery using non-Gaussian mixtures , 2004, SPIE Remote Sensing.

[9]  S. J. Sutley,et al.  Ground-truthing AVIRIS mineral mapping at Cuprite, Nevada , 1992 .

[10]  Fatih Murat Porikli,et al.  Pedestrian Detection via Classification on Riemannian Manifolds , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Stanley R. Rotman,et al.  Segmentation of hyperspectral images from the histograms of principle components , 2002, SPIE Optics + Photonics.

[12]  Luísa Castro,et al.  Hierarchical clustering of multispectral images using combined spectral and spatial criteria , 2005, IEEE Geoscience and Remote Sensing Letters.

[13]  Lorenzo Bruzzone,et al.  Automatic Spectral Rule-Based Preliminary Mapping of Calibrated Landsat TM and ETM+ Images , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[14]  Ujjwal Maulik,et al.  Unsupervised Pixel Classification in Satellite Imagery Using Multiobjective Fuzzy Clustering Combined With SVM Classifier , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Fatih Murat Porikli,et al.  Region Covariance: A Fast Descriptor for Detection and Classification , 2006, ECCV.

[16]  G. Mercier,et al.  Hyperspectral image segmentation with Markov chain model , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).