Fuzzy learning vector quantization for hyperspectral coastal vegetation classification

Abstract Artificial neural networks (ANNs) may be of significant value in extracting vegetation type information in complex vegetation mapping problems, particularly in coastal wetland environments. Unsupervised, self-organizing ANNs have not been employed as frequently as supervised ANNs for vegetation mapping tasks, and further remote sensing research involving fuzzy ANNs is also needed. In this research, the utility of a fuzzy unsupervised ANN, specifically a fuzzy learning vector quantization (FLVQ) ANN, was investigated in the context of hyperspectral AVIRIS image classification. One key feature of the neural approach is that unlike conventional hyperspectral data processing methods, endmembers for a given scene, which can be difficult to determine with confidence, are not required for neural analysis. The classification accuracy of FLVQ was comparable to a conventional supervised multi-layer perceptron, trained with backpropagation (MLP) (KHAT ( Kˆ ) accuracy: 82.82% and 84.66%, respectively; normalized accuracy: 74.60% and 75.85%, respectively), with no significant difference at the 95% confidence level. All neural algorithms in the experiment yielded significantly higher classification accuracies than the conventional endmember-based hyperspectral mapping method assessed (i.e., matched filtering, where Kˆ accuracy = 61.00% and normalized accuracy = 57.96%). FLVQ was also dramatically more computationally efficient than the baseline supervised and unsupervised ANN algorithms tested, including the MLP and the Kohonen self-organizing map (SOM), respectively. The 400-neuron FLVQ network required only 3.6% of the computation time used by the MLP network, and only 5.9% of the MLP time was used by the 588-neuron FLVQ network. In addition, the 400-neuron FLVQ used only 16.7% of the time used by the 400-neuron SOM for model development.

[1]  Helge J. Ritter,et al.  Neural computation and self-organizing maps - an introduction , 1992, Computation and neural systems series.

[2]  Alan H. Strahler,et al.  Fuzzy Neural Network Classification of Global Land Cover from a 1° AVHRR Data Set , 1999 .

[3]  P. Switzer,et al.  A transformation for ordering multispectral data in terms of image quality with implications for noise removal , 1988 .

[4]  Nicolaos B. Karayiannis,et al.  Fuzzy algorithms for learning vector quantization , 1996, IEEE Trans. Neural Networks.

[5]  Palma Blonda,et al.  Model transitions in descending FLVQ , 1998, IEEE Trans. Neural Networks.

[6]  Chein-I Chang,et al.  Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach , 1994, IEEE Trans. Geosci. Remote. Sens..

[7]  Jon Atli Benediktsson,et al.  Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data , 1993 .

[8]  Giles M. Foody,et al.  Cross-entropy for the evaluation of the accuracy of a fuzzy land cover classification with fuzzy ground data , 1995 .

[9]  Philip A. Townsend,et al.  A Quantitative Fuzzy Approach to Assess Mapped Vegetation Classifications for Ecological Applications , 2000 .

[10]  John R. Jensen,et al.  Introductory Digital Image Processing: A Remote Sensing Perspective , 1986 .

[11]  Paul E. Johnson,et al.  Spectral mixture modeling: A new analysis of rock and soil types at the Viking Lander 1 Site , 1986 .

[12]  A. Skidmore,et al.  Spectral discrimination of vegetation types in a coastal wetland , 2003 .

[13]  James C. Bezdek,et al.  An integrated approach to fuzzy learning vector quantization and fuzzy c-means clustering , 1997, IEEE Trans. Fuzzy Syst..

[14]  Freek D. van der Meer,et al.  Calibration of airborne visible - infrared imaging spectrometer data, AVIRIS to reflectance and mineral mapping in hydrothermal alteration zones : an example from the Cuprite mining district , 1994 .

[15]  S. Tompkins,et al.  Optimization of endmembers for spectral mixture analysis , 1997 .

[16]  Lieven Verbeke,et al.  Sub-pixel mapping and sub-pixel sharpening using neural network predicted wavelet coefficients , 2004 .

[17]  C R Hall,et al.  Environmental monitoring and research at the John F. Kennedy Space Center. , 1992, The Journal of the Florida Medical Association.

[18]  Paul A. Schmalzer,et al.  Effects of fire on composition, biomass, and nutrients in oak scrub vegetation on John F. Kennedy Space Center, Florida , 1987 .

[19]  David R. Breininger,et al.  Effects of Fire and Disturbance on Plants and Birds in a Florida Oak/ Palmetto Scrub Community , 1990 .

[20]  Bernd Fritzke,et al.  A Growing Neural Gas Network Learns Topologies , 1994, NIPS.

[21]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[22]  Charles M. Bachmann,et al.  Automatic classification of land cover on Smith Island, VA, using HyMAP imagery , 2002, IEEE Trans. Geosci. Remote. Sens..

[23]  David A. Landgrebe,et al.  Decision boundary feature extraction for nonparametric classification , 1993, IEEE Trans. Syst. Man Cybern..

[24]  Robert De Wulf,et al.  Land cover mapping at sub-pixel scales using linear optimization techniques , 2002 .

[25]  James C. Bezdek,et al.  Two soft relatives of learning vector quantization , 1995, Neural Networks.

[26]  J. Boardman.  Automating spectral unmixing of AVIRIS data using convex geometry concepts , 1993 .

[27]  Ursula C. Benz,et al.  Supervised fuzzy analysis of single- and multichannel SAR data , 1999, IEEE Trans. Geosci. Remote. Sens..

[28]  James C. Bezdek,et al.  Fuzzy Kohonen clustering networks , 1994, Pattern Recognit..

[29]  Didier Tanré,et al.  Second Simulation of the Satellite Signal in the Solar Spectrum, 6S: an overview , 1997, IEEE Trans. Geosci. Remote. Sens..

[30]  Pascal Matsakis,et al.  Evaluation of Fuzzy Partitions , 2000 .

[31]  R. E. Roger.  Sparse inverse covariance matrices and efficient maximum likelihood classification of hyperspectral data , 1996 .

[32]  James C. Bezdek,et al.  Image segmentation using fuzzy LVQ clustering networks , 1992 .

[33]  W. Malkmus,et al.  Random Lorentz band model with exponential-tailed S-1 line-intensity distribution function , 1967 .

[34]  C. S. George Lee,et al.  Neural fuzzy systems: a neuro-fuzzy synergism to intelligent systems , 1996 .

[35]  David R. Breininger,et al.  Mapping Florida Scrub Jay habitat for purposes of land-use management , 1991 .

[36]  R. Lucas,et al.  Non-linear mixture modelling without end-members using an artificial neural network , 1997 .

[37]  William Salas,et al.  A multivariable approach for mapping sub-pixel land cover distributions using MISR and MODIS: Application in the Brazilian Amazon region , 2003 .

[38]  J.A. Benediktsson,et al.  Classification of AVIRIS data from Iceland-geological applications , 1994, Proceedings of IGARSS '94 - 1994 IEEE International Geoscience and Remote Sensing Symposium.

[39]  Stephen V. Stehman,et al.  A Critical Evaluation of the Normalized Error Matrix in Map Accuracy Assessment , 2004 .

[40]  Lalit Kumar,et al.  Mapping Coastal Vegetation Using an Expert System and Hyperspectral Imagery , 2004 .

[41]  Robert A. Schowengerdt,et al.  A review and analysis of backpropagation neural networks for classification of remotely-sensed multi-spectral imagery , 1995 .

[42]  J. C. Price.  How unique are spectral signatures , 1994 .

[43]  G. M. Foody,et al.  Relating the land-cover composition of mixed pixels to artificial neural network classification outpout , 1996 .

[44]  Yang Hong,et al.  A back - propagation neural network for mineralogical mapping from AVIRIS data , 1997 .

[45]  Paul A. Schmalzer,et al.  Geology, geohydrology, and soils of NASA, Kennedy Space Center: A review , 1990 .

[46]  E. Vermote,et al.  Second Simulation Of The Satellite Signal In The Solar Spectrum - 6s Code , 1990, 10th Annual International Symposium on Geoscience and Remote Sensing.

[47]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.