Classification of High Spatial Resolution Imagery Using Improved Gaussian Markov Random-Field-Based Texture Features

Gaussian Markov random fields (GMRFs) are used to analyze textures. GMRFs measure the interdependence of neighboring pixels within a texture to produce features. In this paper, neighboring pixels are taken into account in a priority sequence according to their distance from the center pixel, and a step-by-step least squares method is proposed to extract a novel set of GMRF texture features, named as PS-GMRF. A complete procedure is first designed to classify texture samples of QuickBird imagery. After texture feature extraction, a subset of PS-GMRF features is obtained by the sequential floating forward-selection method. Then, the maximum a posteriori iterated conditional mode classification algorithm is used, involving the selected PS-GMRF texture features in combination with spectral features. The experimental results show that the performance of classifying texture samples on high spatial resolution QuickBird satellite imagery is improved when texture features and spectral features are used jointly, and PS-GMRF features have a higher discrimination power compared to the classical GMRF features, making a notable improvement in classification accuracy from 71.84% to 94.01%. On the other hand, it is found that one of the PS-GMRF texture features - the lowest order variance - is effective for residential-area detection. Some results for IKONOS and SPOT-5 images show that the integration of the lowest order variance with spectral features improves the classification accuracy compared to classification with purely spectral features

[1]  Hichem Frigui,et al.  Fuzzy and possibilistic shell clustering algorithms and their application to boundary detection and surface approximation. II , 1995, IEEE Trans. Fuzzy Syst..

[2]  Meritxell Bach Cuadra,et al.  Region-based satellite image classification: method and validation , 2005, IEEE International Conference on Image Processing 2005.

[3]  John A. Richards,et al.  Efficient maximum likelihood classification for imaging spectrometer data sets , 1994, IEEE Trans. Geosci. Remote. Sens..

[4]  Mihai Datcu,et al.  Spatial information retrieval from remote-sensing images. I. Information theoretical perspective , 1998, IEEE Trans. Geosci. Remote. Sens..

[5]  David A. Clausi,et al.  Design-based texture feature fusion using Gabor filters and co-occurrence probabilities , 2005, IEEE Transactions on Image Processing.

[6]  Rama Chellappa,et al.  Multiresolution Gauss-Markov random field models for texture segmentation , 1997, IEEE Trans. Image Process..

[7]  Mineichi Kudo,et al.  Comparison of algorithms that select features for pattern classifiers , 2000, Pattern Recognit..

[8]  Waqar Ahmad,et al.  A COMPARISON OF OBJECT-ORIENTED AND PIXEL-BASED CLASSIFICATION METHODS FOR MAPPING LAND COVER IN NORTHERN AUSTRALIA. , 2005 .

[9]  Russell G. Congalton,et al.  Classifying and Mapping Forest Cover Types Using Ikonos Imagery in the Northeastern United States , 2002 .

[10]  J. Besag On the Statistical Analysis of Dirty Pictures , 1986 .

[11]  Rama Chellappa,et al.  A model-based approach for estimation of two-dimensional maximum entropy power spectra , 1985, IEEE Trans. Inf. Theory.

[12]  Tim J. Dennis,et al.  Simulated annealing and iterated conditional modes with selective and confidence enhanced update schemes , 1992, [1992] Proceedings Fifth Annual IEEE Symposium on Computer-Based Medical Systems.

[13]  N. Balram,et al.  Noncausal Gauss Markov random fields: Parameter structure and estimation , 1993, IEEE Trans. Inf. Theory.

[14]  J. M. Hans du Buf,et al.  A review of recent texture segmentation and feature extraction techniques , 1993 .

[15]  David A. Clausi,et al.  Comparing cooccurrence probabilities and Markov random fields for texture analysis of SAR sea ice imagery , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[16]  Qi Chen,et al.  Automatic variogram parameter extraction for textural classification of the panchromatic IKONOS imagery , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[17]  Hichem Frigui,et al.  Fuzzy and possibilistic shell clustering algorithms and their application to boundary detection and surface approximation. II , 1995, IEEE Trans. Fuzzy Syst..

[18]  Marvin E. Bauer,et al.  Integrating Contextual Information with per-Pixel Classification for Improved Land Cover Classification , 2000 .

[19]  K. Price,et al.  Optimal Landsat TM band combinations and vegetation indices for discrimination of six grassland types in eastern Kansas , 2002 .

[20]  R. Hall,et al.  Incorporating texture into classification of forest species composition from airborne multispectral images , 2000 .

[21]  Anil K. Jain,et al.  Algorithms for feature selection: An evaluation , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[22]  Alexandre Jouan,et al.  Gabor vs. GMRF features for SAR imagery classification , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[23]  Josef Kittler,et al.  Floating search methods in feature selection , 1994, Pattern Recognit. Lett..

[24]  Richard C. Dubes,et al.  Performance evaluation for four classes of textural features , 1992, Pattern Recognit..

[25]  Brandt Tso,et al.  Scene classification using combined spectral, textural, and contextual information , 2004, SPIE Defense + Commercial Sensing.

[26]  Rama Chellappa,et al.  Classification of textures using Gaussian Markov random fields , 1985, IEEE Trans. Acoust. Speech Signal Process..

[27]  David A. Clausi,et al.  Operational segmentation and classification of SAR sea ice imagery , 2003, IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, 2003.

[28]  Rama Chellappa,et al.  Estimation and choice of neighbors in spatial-interaction models of images , 1983, IEEE Trans. Inf. Theory.

[29]  Anil K. Jain,et al.  Markov Random Field Texture Models , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  Mihai Datcu,et al.  Spatial information retrieval from remote-sensing images. II. Gibbs-Markov random fields , 1998, IEEE Trans. Geosci. Remote. Sens..