Remote Sensing Satellite Images Classification Using Support Vector Machine and Particle Swarm Optimization

This paper introduces a classification system for remote sensing ASTER satellite imagery using SVM and particle swarm optimization (PSO) algorithm. The proposed system starts with the identification of selected area of study. This is followed by a pre-processing phase using mapping polynomial algorithm as geometric correction. Followed by, applying threshold algorithm for image segmentation. Then features are extracted using object based algorithm. Followed by, image classification using SVM and particle swarm optimization(PSO). The PSO is employed as a fast global optimization algorithm instead of using traditional algorithm such as Karush-Kuhn-Tucker conditions. It is implemented and evaluated on real two selected area of interest in the North-Eastern part of the Eastern Desert of Egypt (Halaib Triangle)and (Wadi Shait). The obtained results carried out that the usage of RBF kernel function has the highest classification accuracy ratio as well as Polynomial kernel function.

[1]  Jiong Yang,et al.  An Approach to Active Spatial Data Mining Based on Statistical Information , 2000, IEEE Trans. Knowl. Data Eng..

[2]  Stanisław Lewiński,et al.  Examples of Object-Oriented Classification Performed on High-Resolution Satellite Images , 2004 .

[3]  F. ELtohamy,et al.  Effect of Ground Control points Location and Distribution on Geometric Correction Accuracy of Remote Sensing Satellite Images , 2009 .

[4]  Michael N. Vrahatis,et al.  Recent approaches to global optimization problems through Particle Swarm Optimization , 2002, Natural Computing.

[5]  A. K. Upadhyay,et al.  Spectral mixture analysis for subpixel classification of coconut , 2006 .

[6]  M. Tech,et al.  Spatial Data Mining using Novel Neural Networks for Soil Image Classification and Processing , 2010 .

[7]  Michael Jasinski,et al.  Estimation of subpixel vegetation density of natural regions using satellite multispectral imagery , 1996, IEEE Trans. Geosci. Remote. Sens..

[8]  Hussein A. Abbass,et al.  A Modified Strategy for the Constriction Factor in Particle Swarm Optimization , 2007, ACAL.

[9]  P. S. Hiremath,et al.  Visualization techniques for data mining of Latur district satellite imagery , 2011, ArXiv.

[10]  T. Malvić,et al.  Application of neural networks in petroleum reservoir lithology and saturation prediction , 2009 .

[11]  P. Vamplew,et al.  A Comparative Study of Various Data Mining Techniques as applied to the Modeling of Landslide Susceptibility on the Bellarine Peninsula, Victoria, Australia , 2010 .

[12]  G. Mercier,et al.  Support vector machines for hyperspectral image classification with spectral-based kernels , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[13]  Daniel B. Carr,et al.  A COMPARISON OF CLASSIFICATION METHODS FOR LARGE IMAGERY DATA SETS , 2002 .

[14]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[15]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[16]  Qihao Weng,et al.  A survey of image classification methods and techniques for improving classification performance , 2007 .

[17]  S. Jyothi,et al.  Soil Classification Using Data Mining Techniques: A Comparative Study , 2011 .

[18]  Xianmin Wang,et al.  Spatial Forecast of Landslides in Three Gorges Based On Spatial Data Mining , 2009, Sensors.

[19]  R. Bhaskaran,et al.  Supervised Classification Performance of Multispectral Images , 2010, ArXiv.