Classification of hyperspectral data from urban areas based on extended morphological profiles

Classification of hyperspectral data with high spatial resolution from urban areas is investigated. A method based on mathematical morphology for preprocessing of the hyperspectral data is proposed. In this approach, opening and closing morphological transforms are used in order to isolate bright (opening) and dark (closing) structures in images, where bright/dark means brighter/darker than the surrounding features in the images. A morphological profile is constructed based on the repeated use of openings and closings with a structuring element of increasing size, starting with one original image. In order to apply the morphological approach to hyperspectral data, principal components of the hyperspectral imagery are computed. The most significant principal components are used as base images for an extended morphological profile, i.e., a profile based on more than one original image. In experiments, two hyperspectral urban datasets are classified. The proposed method is used as a preprocessing method for a neural network classifier and compared to more conventional classification methods with different types of statistical computations and feature extraction.

[1]  David A. Landgrebe,et al.  Covariance Matrix Estimation and Classification With Limited Training Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Bor-Chen Kuo,et al.  Improved statistics estimation and feature extraction for hyperspectral data classification , 2001 .

[3]  J.A. Benediktsson,et al.  Morphological transformations and feature extraction of urban data with high spectral and spatial resolution , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[4]  David A. Landgrebe,et al.  Toward an optimal supervised classifier for the analysis of hyperspectral data , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[5]  Jon Atli Benediktsson,et al.  A new approach for the morphological segmentation of high-resolution satellite imagery , 2001, IEEE Trans. Geosci. Remote. Sens..

[6]  David A. Landgrebe,et al.  Feature Extraction Based on Decision Boundaries , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  David A. Landgrebe,et al.  A model-based mixture-supervised classification approach in hyperspectral data analysis , 2002, IEEE Trans. Geosci. Remote. Sens..

[8]  Karl Segl,et al.  Analysis of spectral signatures of urban surfaces for their identification using hyperspectral HyMap data , 2001, IEEE/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas (Cat. No.01EX482).

[9]  José Crespo,et al.  Theoretical aspects of morphological filters by reconstruction , 1995, Signal Process..

[10]  B. Datt,et al.  Sub-pixel analysis of urban surface materials. A case study of Perth, W. Australia , 2000, IGARSS 2000. IEEE 2000 International Geoscience and Remote Sensing Symposium. Taking the Pulse of the Planet: The Role of Remote Sensing in Managing the Environment. Proceedings (Cat. No.00CH37120).

[11]  Laurent Najman,et al.  Definition and some properties of the watershed of a continuous function , 1993 .

[12]  David A. Landgrebe,et al.  The effect of unlabeled samples in reducing the small sample size problem and mitigating the Hughes phenomenon , 1994, IEEE Trans. Geosci. Remote. Sens..

[13]  Pierre Soille,et al.  Advances in mathematical morphology applied to geoscience and remote sensing , 2002, IEEE Trans. Geosci. Remote. Sens..

[14]  Jean Serra,et al.  Image Analysis and Mathematical Morphology , 1983 .

[15]  L. Fiumi.  Evaluation of MIVIS hyperspectral data for mapping covering materials , 2001, IEEE/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas (Cat. No.01EX482).

[16]  Jon Atli Benediktsson,et al.  Image Segmentation Based on the Derivative of the Morphological Profile , 2000, ISMM.

[17]  Bor-Chen Kuo,et al.  A covariance estimator for small sample size classification problems and its application to feature extraction , 2002, IEEE Trans. Geosci. Remote. Sens..

[18]  Pierre Soille,et al.  Morphological Image Analysis: Principles and Applications , 2003 .

[19]  Hermann Kaufmann,et al.  Automated differentiation of urban surfaces based on airborne hyperspectral imagery , 2001, IEEE Trans. Geosci. Remote. Sens..

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

[21]  David A. Landgrebe,et al.  Decision boundary feature extraction for neural networks , 1997, IEEE Trans. Neural Networks.

[22]  A. Plaza,et al.  A new method for target detection in hyperspectral imagery based on extended morphological profiles , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[23]  Jon Atli Benediktsson,et al.  Classification and feature extraction for remote sensing images from urban areas based on morphological transformations , 2003, IEEE Trans. Geosci. Remote. Sens..

[24]  P. Gamba,et al.  Exploiting spectral and spatial information for classifying hyperspectral data in urban areas , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[25]  Bor-Chen Kuo,et al.  A robust classification procedure based on mixture classifiers and nonparametric weighted feature extraction , 2002, IEEE Trans. Geosci. Remote. Sens..

[26]  Serge Beucher,et al.  Use of watersheds in contour detection , 1979 .

[27]  Pavel Pudil,et al.  Introduction to Statistical Pattern Recognition , 2006 .

[28]  Philippe Salembier,et al.  Connected operators and pyramids , 1993, Optics & Photonics.

[29]  Keinosuke Fukunaga,et al.  Introduction to statistical pattern recognition (2nd ed.) , 1990 .