Multiobjective PSO based adaption of neural network topology for pixel classification in satellite imagery

We present the work on efficient classification of multispectral images using soft computing approach.Selection of most discriminative spectral bands and determination of the number of hidden layer neurons are the two most critical issues.We proposed a new multiobjective particle swarm optimization based methodology for adaption of neural network structure for pixel classification of Satellite Imagery.It simultaneously estimates the most discriminative spectral features and the optimal number of nodes in hidden layer.Xie-Beni and β indexes of proposed algorithm are better than MLC and Euclidean Classifier. The proposed work involves the multiobjective PSO based adaption of optimal neural network topology for the classification of multispectral satellite images. It is per pixel supervised classification using spectral bands (original feature space). This paper also presents a thorough experimental analysis to investigate the behavior of neural network classifier for given problem. Based on 1050 number of experiments, we conclude that following two critical issues needs to be addressed: (1) selection of most discriminative spectral bands and (2) determination of optimal number of nodes in hidden layer. We propose new methodology based on multiobjective particle swarm optimization (MOPSO) technique to determine discriminative spectral bands and the number of hidden layer node simultaneously. The accuracy with neural network structure thus obtained is compared with that of traditional classifiers like MLC and Euclidean classifier. The performance of proposed classifier is evaluated quantitatively using Xie-Beni and β indexes. The result shows the superiority of the proposed method to the conventional one.

[1]  Teresa Bernarda Ludermir,et al.  A multi-objective memetic and hybrid methodology for optimizing the parameters and performance of artificial neural networks , 2010, Neurocomputing.

[2]  Athanasios V. Vasilakos,et al.  Comparison of computational intelligence based classification techniques for remotely sensed optical image classification , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[3]  Stefan Roth,et al.  Applications of multi-objective structure optimization , 2006, ESANN.

[4]  B. Mannan,et al.  Fuzzy ARTMAP supervised classification of multi-spectral remotely-sensed images , 1998 .

[5]  Antonio J. Plaza,et al.  On the use of small training sets for neural network-based characterization of mixed pixels in remotely sensed hyperspectral images , 2009, Pattern Recognit..

[6]  Russell C. Eberhart,et al.  A discrete binary version of the particle swarm algorithm , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[7]  Ujjwal Maulik,et al.  Multiobjective Genetic Clustering for Pixel Classification in Remote Sensing Imagery , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[8]  C.A. Coello Coello,et al.  MOPSO: a proposal for multiple objective particle swarm optimization , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[9]  Ashish Ghosh,et al.  Supervised and unsupervised landuse map generation from remotely sensed images using ant based systems , 2011, Appl. Soft Comput..

[10]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[11]  Sultan Noman Qasem,et al.  Multi-objective hybrid evolutionary algorithms for radial basis function neural network design , 2012, Knowl. Based Syst..

[12]  David A. Landgrebe,et al.  Signal Theory Methods in Multispectral Remote Sensing , 2003 .

[13]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[14]  Mausumi Acharyya,et al.  Segmentation of remotely sensed images using wavelet features and their evaluation in soft computing framework , 2003, IEEE Trans. Geosci. Remote. Sens..

[15]  Chi Hau Chen,et al.  Statistical pattern recognition in remote sensing , 2008, Pattern Recognit..

[16]  Okan K. Ersoy,et al.  Border Vector Detection and Adaptation for Classification of Multispectral and Hyperspectral Remote Sensing Images , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[17]  P. Swain,et al.  Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing Data , 1990 .

[18]  D. Fairbairn,et al.  Artificial neural networks for mapping regional‐scale upland vegetation from high spatial resolution imagery , 2006 .

[19]  Saroj K. Meher,et al.  Wavelet-Feature-Based Classifiers for Multispectral Remote-Sensing Images , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[20]  Moncef Gabbouj,et al.  Evolutionary artificial neural networks by multi-dimensional particle swarm optimization , 2009, Neural Networks.

[21]  Carlos A. Coello Coello,et al.  Handling multiple objectives with particle swarm optimization , 2004, IEEE Transactions on Evolutionary Computation.

[22]  Daniel Rivero,et al.  Generation and simplification of Artificial Neural Networks by means of Genetic Programming , 2010, Neurocomputing.

[23]  P. Atkinson,et al.  Introduction Neural networks in remote sensing , 1997 .

[24]  G. Wilkinson,et al.  Training strategies for neural network soft classification of remotely-sensed imagery , 1997 .

[25]  Fuqing Zhao,et al.  Application of An Improved Particle Swarm Optimization Algorithm for Neural Network Training* , 2005, 2005 International Conference on Neural Networks and Brain.

[26]  Jonathan E. Fieldsend,et al.  Pareto evolutionary neural networks , 2005, IEEE Transactions on Neural Networks.

[27]  Richard J. Duro,et al.  Gaussian synapse ANNs in multi- and hyperspectral image data analysis , 2003, IEEE Trans. Instrum. Meas..

[28]  Okan K. Ersoy,et al.  A statistical self-organizing learning system for remote sensing classification , 2005, IEEE Transactions on Geoscience and Remote Sensing.

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

[30]  Farid Melgani,et al.  Clustering of Hyperspectral Images Based on Multiobjective Particle Swarm Optimization , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[31]  Farid Melgani,et al.  Semisupervised PSO-SVM Regression for Biophysical Parameter Estimation , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[32]  Teresa Bernarda Ludermir,et al.  An Analysis Of PSO Hybrid Algorithms For Feed-Forward Neural Networks Training , 2006, 2006 Ninth Brazilian Symposium on Neural Networks (SBRN'06).

[33]  Paul M. Mather,et al.  Computer processing of remotely-sensed images , 2016 .

[34]  Giles M. Foody,et al.  An evaluation of some factors affecting the accuracy of classification by an artificial neural network , 1997 .

[35]  Eric Granger,et al.  Evolving ARTMAP neural networks using Multi-Objective Particle Swarm Optimization , 2010, IEEE Congress on Evolutionary Computation.