Mapping a specific class for priority habitats monitoring from satellite sensor data

The European Union's Habitats Directive aims to protect biodiversity through the conservation of habitats. If a habitat of interest corresponds to spectrally separable land cover class(es), then this activity can benefit from the production of accurate land cover maps from remotely sensed imagery. Traditionally, the image classification techniques used assume that the set of classes has been defined exhaustively, which requires all of the classes in the region to be included explicitly in the analysis. Often, however, interest focuses on just one or a small sub‐set of the classes occurring in the region that represent the habitat(s) of particular interest. Moreover, given that the size of a training set required for an image classification is typically taken to be a function of the number of classes and discriminating variables (e.g. wavebands) used in the classification, the satisfaction of the assumption of an exhaustively defined set of classes requires that much effort is directed wastefully on classes of little, if any, direct interest. Savings in training could be achieved by focusing on the class(es) of specific interest. A more appropriate approach for mapping a specific class may be to adopt a binary classification analysis that simply seeks to separate the class of interest from all others. In this way the analysis focuses on the class(es) of interest and a small training set may be used. An attractive means to achieve this is through the adoption of decision tree‐ and support vector machine‐based approaches to classification. This paper evaluates the accuracy with which a habitat of interest to the EU Habitats Directive, fen, can be mapped from Landsat ETM+ imagery of the Norfolk Broads using such classifiers as well as, for comparative purposes, a standard maximum likelihood decision rule implemented by a discriminant analysis. All analyses yielded accurate classifications, with a conventional approach based on a maximum likelihood allocation providing an overall classification accuracy of 88.4%. However, both the decision tree‐ and support vector machine‐based approaches provided classifications that were significantly more accurate than conventional maximum likelihood classification (p<0.05), with overall accuracies of 91.6 and 93.6%, respectively (table 3). The results highlight the ability to focus the analysis on the class of interest in a manner that is less wasteful of resources and effort and that yields a more accurate classification than the standard approach.

[1]  S. E. Hampson,et al.  Linear function neurons: Structure and training , 1986, Biological Cybernetics.

[2]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[3]  Giles M. Foody,et al.  Changing land cover , 2004 .

[4]  R. Reynolds,et al.  A non-parametric, supervised classification of vegetation types on the Kaibab National Forest using decision trees , 2003 .

[5]  Vladimir Naumovich Vapni.  The Nature of Statistical Learning Theory , 1995 .

[6]  Martin Brown,et al.  Support vector machines for optimal classification and spectral unmixing , 1999 .

[7]  A. Jawad,et al.  Computer processing of remotely sensed images , 2005 .

[8]  Giles M. Foody,et al.  Supervised image classification by MLP and RBF neural networks with and without an exhaustively defined set of classes , 2004 .

[9]  S. Gunn.  Support Vector Machines for Classification and Regression , 1998 .

[10]  Paul M. Mather,et al.  An assessment of the effectiveness of decision tree methods for land cover classification , 2003 .

[11]  G. Foody.  Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy , 2004 .

[12]  C. Brodley,et al.  Decision tree classification of land cover from remotely sensed data , 1997 .

[13]  Vladimir Cherkassky,et al.  Learning from Data: Concepts, Theory, and Methods , 1998 .

[14]  G. Rajashekar,et al.  Diagnostic analysis of conservation zones using remote sensing and GIS techniques in wet evergreen forests of the Western Ghats – An ecological hotspot, Tamil Nadu, India , 2003, Biodiversity & Conservation.

[15]  B. Schölkopf,et al.  Advances in kernel methods: support vector learning , 1999 .

[16]  Sharon K. Collinge,et al.  Ecological consequences of habitat fragmentation: implications for landscape architecture and planning , 1996 .

[17]  Walter V. Reid,et al.  Conserving the World's Biological Diversity , 1990 .

[18]  David A. Landgrebe,et al.  A survey of decision tree classifier methodology , 1991, IEEE Trans. Syst. Man Cybern..

[19]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[20]  Paul M. Mather,et al.  Support vector machines for classification in remote sensing , 2005 .

[21]  R. M. Lark,et al.  Components of accuracy of maps with special reference to discriminant analysis on remote sensor data , 1995 .

[22]  L. S. Davis,et al.  An assessment of support vector machines for land cover classi(cid:142) cation , 2002 .

[23]  Andrew Jordan,et al.  Implementing EU biodiversity policy: UK experiences , 2000 .

[24]  Mathieu Rouget,et al.  Measuring conservation value at fine and broad scales: implications for a diverse and fragmented region, the Agulhas Plain , 2003 .

[25]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[26]  Giles M. Foody,et al.  A relative evaluation of multiclass image classification by support vector machines , 2004, IEEE Transactions on Geoscience and Remote Sensing.

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

[28]  Erwin H. Bulte,et al.  Conserving Nature's Diversity: Insights from Biology, Ethics and Economics: Insights from Biology, Ethics and Economics , 2000 .

[29]  Giles M. Foody,et al.  Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification , 2004 .

[30]  R. Congalton,et al.  Accuracy assessment: a user's perspective , 1986 .

[31]  L. Zhang,et al.  Comparison of Neural Networks and Statistical Methods in Classification of Ecological Habitats Using FIA Data , 2003 .

[32]  M. Aizerman,et al.  Theoretical Foundations of the Potential Function Method in Pattern Recognition Learning , 1964 .

[33]  Maggi Kelly,et al.  Support vector machines for predicting distribution of Sudden Oak Death in California , 2005 .

[34]  J. Kerr,et al.  From space to species: ecological applications for remote sensing , 2003 .

[35]  E Brown de Colstoun,et al.  National Park vegetation mapping using multitemporal Landsat 7 data and a decision tree classifier , 2003 .

[36]  Roy Brouwer,et al.  Ecosystem services value, research needs, and policy relevance: a commentary , 1998 .

[37]  Giles M. Foody,et al.  IDENTIFICATION OF SPECIFIC TREE SPECIES IN ANCIENT SEMI-NATURAL WOODLAND FROM DIGITAL AERIAL SENSOR IMAGERY , 2005 .

[38]  A. Cooper,et al.  Sampling Broad Habitat change to assess biodiversity conservation action in Northern Ireland. , 2003, Journal of environmental management.

[39]  A. Pullin,et al.  Do conservation managers use scientific evidence to support their decision-making? , 2004 .

[40]  Ray Bert,et al.  Book Review: Computer Processing of Remotely-Sensed Images: An Introduction, Third Edition , by Paul M. Mather. Chichester, United Kingdom: John Wiley & Sons Ltd., 2004 , 2004 .

[41]  Yunqian Ma,et al.  Practical selection of SVM parameters and noise estimation for SVM regression , 2004, Neural Networks.

[42]  Michael Stocking,et al.  People and environment , 1995 .

[43]  S. Prasher,et al.  Classification of hyperspectral data by decision trees and artificial neural networks to identify weed stress and nitrogen status of corn , 2003 .

[44]  Sanjay Tomar,et al.  Biodiversity characterization at landscape level using geospatial modelling technique , 2000 .