CHANGE DETECTION SOFTWARE USING SELF-ORGANIZING FEATURE MAPS

Os mapas auto-organizaveis (SOFM) consistem em um tipo de rede neural artificial que permite a conversao de dados de alta dimensao, complexos e nao lineares, em simples relacoes geometricas com baixa dimensionalidade. Este metodo tambem pode ser utilizado para a classificacao de imagens de sensoriamento remoto, pois permite a compressao de dados de alta dimensao preservando as relacoes topologicas dos dados primarios. Este trabalho objetiva desenvolver uma metodologia eficaz para a utilizacao de mapas auto-organizaveis na deteccao de mudancas. No presente estudo o SOFM e utilizado para a classificacao nao supervisionada de dados de sensoriamento remoto, considerando os seguintes atributos: espaciais (x, y), espectrais e temporais. O metodo e empregado na regiao oeste da Bahia, que teve recentemente um aumento significativo em monoculturas. Testes foram realizados com os parâmetros do SOFM com o objetivo de refinar o mapa de deteccao demudancas. O SOFM possibilita uma melhor selecao de celulas e dos correspondentes vetores de peso, que mostram o processo de ordenacao e agrupamento hierarquicodos dados. Esta informacao e essencial para identificar mudancas ao longo do tempo. Um programa em linguagem C ++ do metodo proposto foi desenvolvido. ABSTRACT . Self-organizing feature maps (SOFM) consist of a type of artificial neural network that allows the conversion from high-dimensional data into simple geometric relationships with low-dimensionality. This method can also be used for classification of remote sensing images because it allows the compression of high dimensional data while preserving the most important topological and metric relationships of the primary data. This paper aims to develop an effective methodology forusing self-organizing maps in change detection. In this study, SOFM is used for unsupervised classification of remote sensing data, considering the following attributes: spatial (x and y), spectral and temporal. The method is tested and simulated in the western region of Bahia that has observed a significant increase in mechanized agriculture. Tests were performed with the SOFM parameters for the purpose of fine tuning a change detection map. The SOFM provides the best selection of cell and corresponding adjustment of weight vectors, which show the process of ordering and hierarchical clustering of the data. This information is essential to identify changes over time. All algorithms were implemented in C++ language. Keywords : unsupervised classification; land cover; multitemporal analysis; remote sensing

[1]  I︠a︡. Z. T︠S︡ypkin,et al.  Foundations of the theory of learning systems , 1973 .

[2]  P. E. Anuta,et al.  Change detection in coastal zone environments , 1977 .

[3]  K. K. Mayo,et al.  Monitoring land-cover change by principal component analysis of multitemporal Landsat data. , 1980 .

[4]  Philip J. Howarth,et al.  Procedures for change detection using Landsat digital data , 1981 .

[5]  E. LeDrew,et al.  Application of principal components analysis to change detection , 1987 .

[6]  T. Kohonen,et al.  Statistical pattern recognition with neural networks: benchmarking studies , 1988, IEEE 1988 International Conference on Neural Networks.

[7]  Ashbindu Singh,et al.  Review Article Digital change detection techniques using remotely-sensed data , 1989 .

[8]  J. Hill,et al.  Radiometric correction of multitemporal Thematic Mapper data for use in agricultural land-cover classification and vegetation monitoring , 1991 .

[9]  S. Goetz,et al.  Radiometric rectification - Toward a common radiometric response among multidate, multisensor images , 1991 .

[10]  Peng Gong,et al.  Registration-noise reduction in difference images for change detection , 1992 .

[11]  J. R. Jensen,et al.  An evaluation of the CoastWatch change detection protocol in South Carolina , 1993 .

[12]  Robert A. Schowengerdt,et al.  A detailed comparison of backpropagation neural network and maximum-likelihood classifiers for urban land use classification , 1995, IEEE Trans. Geosci. Remote. Sens..

[13]  Vicente P. Soares,et al.  Eucalyptus forest change classification using multi-date Landsat TM data , 1995, Remote Sensing.

[14]  S. Gopal,et al.  Remote sensing of forest change using artificial neural networks , 1996, IEEE Trans. Geosci. Remote. Sens..

[15]  Gail A. Carpenter,et al.  ART neural networks for remote sensing: vegetation classification from Landsat TM and terrain data , 1997, IEEE Trans. Geosci. Remote. Sens..

[16]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[17]  Siamak Khorram,et al.  The effects of image misregistration on the accuracy of remotely sensed change detection , 1998, IEEE Trans. Geosci. Remote. Sens..

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

[19]  M. Winograd,et al.  Dynamics of the agricultural frontier in the Amazon and savannas of Brazil: analyzing the impact of policy and technology , 1998 .

[20]  C. Elvidge,et al.  NALC Land Cover Change Detection Pilot Study , 1998 .

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

[22]  E. Lambin.  Monitoring forest degradation in tropical regions by remote sensing: some methodological issues , 1999 .

[23]  S. Khorram,et al.  Remotely Sensed Change Detection Based on Artificial Neural Networks , 1999 .

[24]  Janet Franklin,et al.  A Neural Network Method for Efficient Vegetation Mapping , 1999 .

[25]  Pol Coppin,et al.  Operational monitoring of green biomass change for forest management , 2001 .

[26]  Steven E. Franklin,et al.  Interpretation and Classification of Partially Harvested Forest Stands in the Fundy Model Forest Using Multitemporal Landsat TM Digital Data , 2000 .

[27]  S. L Furby,et al.  Calibrating images from different dates to ‘like-value’ digital counts , 2001 .

[28]  Yong Du,et al.  Radiometric normalization, compositing, and quality control for satellite high resolution image mosaics over large areas , 2001, IEEE Trans. Geosci. Remote. Sens..

[29]  S. Sader,et al.  Comparison of change-detection techniques for monitoring tropical forest clearing and vegetation regrowth in a time series , 2001 .

[30]  Steven A. Sader,et al.  Forest change monitoring of a remote biosphere reserve , 2001 .

[31]  D. Kaimowitz,et al.  Soybean Technology and the Loss of Natural Vegetation in Brazil and Bolivia , 2001 .

[32]  V. Radeloff,et al.  Phenological differences in Tasseled Cap indices improve deciduous forest classification , 2002 .

[33]  S. Sader,et al.  Detection of forest harvest type using multiple dates of Landsat TM imagery , 2002 .

[34]  S. Franklin,et al.  Large-area forest structure change detection: An example , 2002 .

[35]  J. Cihlar,et al.  Radiometric normalization of multitemporal high-resolution satellite images with quality control for land cover change detection , 2002 .

[36]  Michael A. Wulder,et al.  Mountain Pine Beetle Red-Attack Forest Damage Classification Using Stratified Landsat TM Data in British Columbia, Canada , 2003 .

[37]  Michael A. Wulder,et al.  Sensitivity of the thematic mapper enhanced wetness difference index to detect mountain pine beetle red-attack damage , 2003 .

[38]  G. Hay,et al.  A Multiscale Object-Specific Approach to Digital Change Detection , 2003 .

[39]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[40]  Stephen Grossberg,et al.  Adaptive pattern classification and universal recoding: II. Feedback, expectation, olfaction, illusions , 1976, Biological Cybernetics.

[41]  M. Canty,et al.  Automatic radiometric normalization of multitemporal satellite imagery , 2004 .

[42]  D. Lu,et al.  Change detection techniques , 2004 .

[43]  Pol Coppin,et al.  Review ArticleDigital change detection methods in ecosystem monitoring: a review , 2004 .

[44]  Ma Jianwen,et al.  Land-use classification using ASTER data and self-organized neutral networks , 2005 .

[45]  Suming Jin,et al.  Comparison of time series tasseled cap wetness and the normalized difference moisture index in detecting forest disturbances , 2005 .

[46]  Hassiba Nemmour,et al.  Neural Network Combination by Fuzzy Integral for Robust Change Detection in Remotely Sensed Imagery , 2005, EURASIP J. Adv. Signal Process..

[47]  M. Bauer,et al.  Land cover classification and change analysis of the Twin Cities (Minnesota) Metropolitan Area by multitemporal Landsat remote sensing , 2005 .

[48]  Renato Fontes Guimarães,et al.  Normalization of Multi-Temporal Images Using a New Change Detection Method Based on the Spectral Classifier , 2006, 2006 IEEE International Symposium on Geoscience and Remote Sensing.

[49]  José A. Sobrino,et al.  Radiometric correction effects in Landsat multi‐date/multi‐sensor change detection studies , 2006 .

[50]  Siamak Khorram,et al.  Correspondence analysis for detecting land cover change , 2006 .

[51]  Michael S. Ramsey,et al.  Radiometric normalization and image mosaic generation of ASTER thermal infrared data: An application to extensive sand sheets and dune fields , 2008 .

[52]  João Paulo Papa,et al.  Robust Pruning of Training Patterns for Optimum-Path Forest Classification Applied to Satellite-Based Rainfall Occurrence Estimation , 2010, IEEE Geoscience and Remote Sensing Letters.

[53]  Christian Jacob,et al.  Self-Organizing Feature Maps , 2010, Encyclopedia of Machine Learning.

[54]  Alan R. Gillespie,et al.  A New Approach to Change Vector Analysis Using Distance and Similarity Measures , 2011, Remote. Sens..

[55]  Elizabeth Maria Feitosa da Rocha de Souza,et al.  SENSORIAMENTO REMOTO NO ESTUDO DA VEGETAÇÃO (MATA ATLÂNTICA): COMPARAÇÕES DAS CURVAS ESPECTRAIS DE SENSORES MULTIESPECTRAIS E HIPERESPECTRAIS , 2012 .