Comparison of Neural Classification Algorithms Applied to Land Cover Mapping

We compared the performance of several supervised classification algorithms on multi-source remotely sensed images. Apart from the Multi-Layer Perceptron, K-Nearest-Neighbour and Radial Basis Function network approaches, we looked more in detail at the Support Vector Machine classifier, which recently showed promising results in our setting. In particular, it is able to provide meaningful answers for the analysis of mixed pixels. They correspond to areas on the ground that comprise more than one distinct class, representing a major challenge for the interpretability of the final land-cover maps. To assess their impact, we performed a rejection-based analysis, allowing classifiers to refuse answers on pixels they can not associate mainly with one class. The experimental results lead to the conclusion that 1vs1 SVM approach with a linear kernel (using Bradley-Terry coupling) has to be preferred over all other classification algorithms examined, both in terms of accuracy as well as ease of visual interpretation.

[1]  Silvia Scarpetta,et al.  Application of neural unsupervised methods to environmental factor analysis of multi-spectral images with texture features , 2008 .

[2]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[3]  Andrew R. Webb,et al.  Statistical Pattern Recognition , 1999 .

[4]  A. Atiya,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2005, IEEE Transactions on Neural Networks.

[5]  Günther Palm,et al.  Comparison of Multiclass SVM Decomposition Schemes for Visual Object Recognition , 2005, DAGM-Symposium.

[6]  C. W. Therrien,et al.  Decision, Estimation and Classification: An Introduction to Pattern Recognition and Related Topics , 1989 .

[7]  J. MacQueen Some methods for classification and analysis of multivariate observations , 1967 .

[8]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[9]  Friedhelm Schwenker,et al.  Fuzzy-Input Fuzzy-Output One-Against-All Support Vector Machines , 2007, KES.

[10]  Lorenzo Bruzzone,et al.  A multiscale expectation-maximization semisupervised classifier suitable for badly posed image classification , 2006, IEEE Transactions on Image Processing.

[11]  L. Breiman,et al.  Variable Kernel Estimates of Multivariate Densities , 1977 .

[12]  Jürgen Schürmann,et al.  Pattern classification - a unified view of statistical and neural approaches , 2008 .

[13]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[14]  Tshilidzi Marwala,et al.  Image Classification Using SVMs: One-against-One Vs One-against-All , 2007, ArXiv.

[15]  Robert Tibshirani,et al.  Classification by Pairwise Coupling , 1997, NIPS.

[16]  R. A. Bradley,et al.  RANK ANALYSIS OF INCOMPLETE BLOCK DESIGNS , 1952 .

[17]  Friedhelm Schwenker,et al.  Three learning phases for radial-basis-function networks , 2001, Neural Networks.

[18]  Paul M. Mather,et al.  Computer Processing of Remotely-Sensed Images: An Introduction , 1988 .

[19]  Lorenzo Bruzzone,et al.  An advanced semi-supervised SVM classifier for the analysis of hyperspectral remote sensing data , 2006, SPIE Remote Sensing.

[20]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[21]  Günther Palm,et al.  Using Dempster-Shafer Theory in MCF Systems to Reject Samples , 2005, Multiple Classifier Systems.

[22]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[23]  R. A. Bradley,et al.  Rank Analysis of Incomplete Block Designs: I. The Method of Paired Comparisons , 1952 .

[24]  Pramod K. Varshney,et al.  Decision tree regression for soft classification of remote sensing data , 2005 .

[25]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[26]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[27]  Cecilio Angulo,et al.  Multi-Classification by Using Tri-Class SVM , 2006, Neural Processing Letters.