A novel approach to the classification of regional-scale Radar mosaics for tropical vegetation mapping

The Global Rain Forest Mapping (GRFM) radar mosaics, generated from L-band Japanese Earth Resources Satellite 1 imagery downsampled to 100-m pixel size, provide a two-season spatially continuous coverage of the humid tropical ecosystems of the world. This paper presents a novel classification approach suitable for regional-scale vegetation mapping using the GRFM datasets. The mapping system consists of: 1) an application-dependent wavelet-based edge-preserving smoothing algorithm and 2) a two-stage per-pixel hybrid learning nearest multiple-prototype (NMP) classifier, whose unsupervised first stage is a per-pixel near-optimal vector quantizer, called enhanced Linde-Buzo-Gray (ELBG), recently proposed in pattern recognition literature. Identified as ENMP (NMP with ELBG), this novel classification approach is compared against two alternative systems in the classification of forest cover disturbances located across an area in the Amazon Basin. Surface classes of interest are primary forest, degraded forest, nonforest, and water bodies. Reference maps, derived from 30-m resolution Landsat Thematic Mapper imagery, are provided by the National Aeronautics and Space Administration and the Food and Agriculture Organization of the United Nations. Abundant quantitative and qualitative evidence shows that: 1) in a forest/nonforest data-mapping task, ENMP provides a testing accuracy of 87%, in line with training accuracies, i.e., the proposed method seems capable of generalizing well over the GRFM South America dataset and 2) among three competing approaches, ENMP provides the best compromise between ease of use, mapping accuracy, and computational time. Starting from these results, ENMP is employed to generate a swamp forest map of the whole Amazon Basin from the two-season GRFM radar mosaic of South America, in the context of the Global Land Cover project (GLC 2000).

[1]  Shigeo Abe DrEng.  Pattern Classification , 2001, Springer London.

[2]  C. Estreguil,et al.  AVHRR for global tropical forest monitoring: The lessons of the TREES project , 1995 .

[3]  Stéphane Mallat,et al.  Singularity detection and processing with wavelets , 1992, IEEE Trans. Inf. Theory.

[4]  S. Saatchi,et al.  Application of multiscale texture in classifying JERS-1 radar data over tropical vegetation , 2002 .

[5]  Investigation of Tropical Rain Forest in Central Amazonia, Brazil based on JERS-1 SAR Images , 1995 .

[6]  A. Rosenqvist,et al.  New perspectives on global ecosystems from wide-area radar mosaics: Flooded forest mapping in the tropics , 2000 .

[7]  Fabio Rocca,et al.  Multibaseline InSAR DEM reconstruction: the wavelet approach , 1999, IEEE Trans. Geosci. Remote. Sens..

[8]  Stéphane Mallat,et al.  A Wavelet Tour of Signal Processing, 2nd Edition , 1999 .

[9]  Marc Simard,et al.  Large-scale vegetation maps derived from the combined L-band GRFM and C-band CAMP wide area radar mosaics of Central Africa , 2002 .

[10]  M. Simard,et al.  Speckle filtering, segmentation and classification of polarimetric SAR data: a unified approach based on the wavelet transform , 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]  G.F. De Grandi,et al.  Segmentation and labeling of polarimetric SAR data: can wavelets help? , 2001, IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217).

[12]  C. Tucker,et al.  Tropical Deforestation and Habitat Fragmentation in the Amazon: Satellite Data from 1978 to 1988 , 1993, Science.

[13]  H. Hirosawa,et al.  Suppression of speckle in synthetic aperture radar images using wavelet , 1998 .

[14]  A. Belward,et al.  GLC2000: a new approach to global land cover mapping from Earth observation data , 2005 .

[15]  G Patane,et al.  Fully automatic clustering system , 2002, IEEE Trans. Neural Networks.

[16]  L.V. Dutra,et al.  Land cover classification based on multi-date JERS-1 imagery as a basis for deforestation detection , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[17]  J. Proctor,et al.  Nature and Dynamics of Forest-Savanna Boundaries. , 1994 .

[18]  James C. Bezdek,et al.  Multiple-prototype classifier design , 1998, IEEE Trans. Syst. Man Cybern. Part C.

[19]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[20]  Shaun Quegan,et al.  ERS-1 observations and potential for use in tropical forest monitoring , 1995, 1995 International Geoscience and Remote Sensing Symposium, IGARSS '95. Quantitative Remote Sensing for Science and Applications.

[21]  Anil K. Jain,et al.  A Clustering Performance Measure Based on Fuzzy Set Decomposition , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  S. Saatchi,et al.  Mapping land cover types in the Amazon Basin using 1 km JERS-1 mosaic , 2000 .

[23]  Qiong Jackson,et al.  An adaptive classifier design for high-dimensional data analysis with a limited training data set , 2001, IEEE Trans. Geosci. Remote. Sens..

[24]  M. Simard,et al.  Classification of JERS-1 image mosaic of Central Africa using a supervised multiscale classifier of texture features , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[25]  Seisuke Fukuda,et al.  Smoothing effect of wavelet-based speckle filtering: the Haar basis case , 1999, IEEE Trans. Geosci. Remote. Sens..

[26]  Steffen Fritz,et al.  A vegetation map of South America , 2002 .

[27]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[28]  G. DeGrandi,et al.  Adaptation of the Wavelet Transform for the Construction of Multiscale Texture Maps of SAR Images , 1998 .

[29]  A. Baraldi,et al.  A vegetation map of the Central Congo basin derived from microwave and optical remote sensing data using a variable resolution classification approach , 2001, IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217).

[30]  M. Simard,et al.  Singularity analysis with wavelets in polarimetric SAR imagery for vegetation mapping applications , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[31]  M. Sgrenzaroli,et al.  Tropical forest cover monitoring: Estimates from the GRFM JERS-1 radar mosaics using wavelet zooming techniques and validation , 2002 .

[32]  M. Simard,et al.  Classification of the Gabon SAR mosaic using a wavelet based rule classifier , 1999, IEEE 1999 International Geoscience and Remote Sensing Symposium. IGARSS'99 (Cat. No.99CH36293).

[33]  Stéphane Mallat,et al.  Characterization of Signals from Multiscale Edges , 2011, IEEE Trans. Pattern Anal. Mach. Intell..

[34]  Paul Siqueira,et al.  A continental-scale mosaic of the Amazon basin using JERS-1 SAR , 2000, IEEE Trans. Geosci. Remote. Sens..

[35]  Sassan Saatchi,et al.  The use of decision tree and multiscale texture for classification of JERS-1 SAR data over tropical forest , 2000, IEEE Trans. Geosci. Remote. Sens..

[36]  Philip H. Swain,et al.  Remote Sensing: The Quantitative Approach , 1981, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[37]  Giuseppe Patanè,et al.  The enhanced LBG algorithm , 2001, Neural Networks.

[38]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[39]  Seisuke Fukuda,et al.  A wavelet-based texture feature set applied to classification of multifrequency polarimetric SAR images , 1999, IEEE Trans. Geosci. Remote. Sens..

[40]  Andreas Niedermeier,et al.  Detection of coastlines in SAR images using wavelet methods , 2000, IEEE Trans. Geosci. Remote. Sens..

[41]  Anil K. Jain,et al.  Algorithms for Clustering Data , 1988 .

[42]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.

[43]  Goze B. Bénié,et al.  Analysis of speckle noise contribution on wavelet decomposition of SAR images , 1998, IEEE Trans. Geosci. Remote. Sens..

[44]  F. Achard,et al.  Determination of Deforestation Rates of the World's Humid Tropical Forests , 2002, Science.

[45]  S. Saatchi,et al.  The Global Rain Forest Mapping project - A review , 2000 .

[46]  K. P. B. Thomson,et al.  The evaluation of segmentation results and the overlapping area matrix , 1997 .

[47]  John A. Richards,et al.  Remote Sensing Digital Image Analysis , 1986 .

[48]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[49]  C. Johnston.  Geographic Information Systems in Ecology , 1998 .

[50]  Palma Blonda,et al.  Contextual clustering for image segmentation , 2000 .

[51]  Giles M. Foody,et al.  Status of land cover classification accuracy assessment , 2002 .

[52]  S. Mallat.  VI – Wavelet zoom , 1999 .

[53]  A. Baraldi,et al.  High-resolution tropical forest mapping of the Amazon basin: a novel classification approach for the GRFM radar mosaic , 2001, IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No.01CH37217).

[54]  Paul M. Mather,et al.  An analysis of deforestation patterns in the extractive reserves of Acre, Amazonia from satellite imagery: A landscape ecological approach , 2000 .

[55]  Sassan Saatchi,et al.  The Global Rain Forest Mapping Project JERS-1 radar mosaic of tropical Africa: development and product characterization aspects , 2000, IEEE Trans. Geosci. Remote. Sens..

[56]  Jean Carletta,et al.  Assessing Agreement on Classification Tasks: The Kappa Statistic , 1996, CL.

[57]  S. Mallat.  A wavelet tour of signal processing , 1998 .