The International Arab Journal of Information Technology (IAJIT)

..............................
..............................
..............................


Assessment of Ensemble Classifiers Using the Bagging

This study evaluates an approach for Land-Use Land-Cover classification (LULC) using multispectral satellite images. This proposed approach uses the Bagging Ensemble (BE) technique with Random Forest (RF) as a base classifier for improving classification performance by reducing errors and prediction variance. A pixel-based supervised classification technique with Principle Component Analysis (PCA) for feature selection from available attributes using a Landsat 8 image is developed. These attributes include coastal, visible, near-infrared, short-wave infrared and thermal bands in addition to Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). The study is performed in a heterogeneous coastal area divided into five classes: water, vegetation, grass-lake-type, sand, and building. To evaluate the classification accuracy of BE with RF, it is compared to BE with Support Vector Machine (SVM) and Neural Network (NN) as base classifiers. The results are evaluated using the following output: commission, omission errors, and overall accuracy. The results showed that the proposed approach using BE with RF outperforms SVM and NN classifiers with 93.3% overall accuracy. The BE with SVM and NN classifiers yielded 92.6% and 92.1% overall accuracy, respectively. It is revealed that using BE with RF as a base classifier outperforms other base classifiers as SVM and NN. In addition, omission and commission errors were reduced by using BE with RF and NN classifiers.


[1] Adam E., Ismail R., and Mutanga O., A Comparison of Selected Machine learning Classifiers in Mapping a South African Heterogeneous Coastal Zone: Testing the Utility of an Object Based Classification with Worldveiw-2 Imagery, in Proceedings of Earth Resources and Environmental Remote Sensing, Edinburgh, 2012.

[2] Akar O. and Gungor O., Classification of Multispectral Images Using Random Forest Algorithm, Geodesy and Geoinformation, vol. 1, no. 2, pp. 105-112, 2012.

[3] Ali E., Impact of Drain Water on Water Quality and Eutrophication Status of Lake Burullus, Egypt, A Southern Mediterranean Lagoon, African Journal of Aquatic Science, vol. 36, no. 3, pp. 267-277, 2011.

[4] Ananth R., The Levenberg-Marquardt Algorithm.8, Available at: http://www.ananth.in/Notes_files/lmtut.pdf, Last Visited 2004.

[5] Behzad S., Predicting the Trend of Land Use Changes Using Artificial Neural Network and Markov Chain Model (Case Study: Kermanshah City), Research Journal of Environmental and Earth Sciences, vol. 6, no. 4, pp. 215-226, 2014.

[6] Benediktsson J., Chanussot J., and Mathieu F., Multiple Classi er Systems in Remote Sensing: From Basics to Recent Developments, Multiple Classifier Systems Lecture Notes in Computer Science, Springer Berlin, pp. 501-512, 2007.

[7] Benjamin L. and Bernd B., Data Mining and Support Vector Regression Machine Learning in Semiconductor Manufacturing to Improve Virtual Metrology, in Proceedings of 46th Hawaii International Conference on System Sciences, Maui, HI, pp. 3447-3456, 2013.

[8] Boukir S., Guo L. and Chehata N., Classification of Remote Sensing Data using Margin-Based Ensemble Methods, in Proceedings of Institute of Electrical and Electronics Engineers International Conference on Image Processing, Melbourne, pp. 2601-2606, 2013.

[9] Breiman L., Bagging Predictors, Machine Learning, vol. 24, no. 2, pp. 123-140, 1996.

[10] Breiman L., Manual on Setting up, using, and Understanding Random Forest V3.1, Available, at:https://www.stat.berkeley.edu/~breiman/Using _random_forests_V3.1.pdf, Last Visited 2015.

[11] Breiman L., Random Forests, Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.

[12] Ceyhun O. and Yalcin A., Remote Sensing of Water Depths in Shallow Waters via Artificial Neural Networks, Estuarine Coastaand Shelf Science, vol. 89, no.1, pp. 89-96, 2010.

[13] Charlie L., Landsat 8 Bands, http://landsat.gsfc.nasa.gov/?page_id=5377, Last Visited 2015.

[14] Chu H. and Ge L., Combination of Genetic Algorithm and Dempster-Shafer Theory of Evidence for Land Cover Classification Using Integration of SAR and Optical Satellite Imagery, International Society for Photogrammetry and Remote Sensing, vol. XXXIX-B7, pp. 173-178, 2012.

[15] Dara R., Cooperative Training in Multiple Classifier Systems Theses, University of Waterloo, 2007.

[16] Du P., Zhang W., and Sun H., Multiple Classifier Combination for Hyperspectral Remote Sensing Image Classification, in Proceedings of 8th International Workshop on Multiple Classifier Systems, Berlin, pp. 52-61, 2009.

[17] Galar M., Fernandez A., Barrenechea E., Bustince H., and Herrera F., A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 42, no. 4, pp. 463-484, 2011.

[18] Ghimire B., Rogan J., Galiano V., Panday P., and Neeti N., An Evaluation of Bagging, Boosting, and Random Forests for Land-Cover Classification in Cape Cod, Massachusetts, USA, GIScience & Remote Sensing, vol. 49, no. 5, pp. 623-643, 2012.

[19] Guan H., Yu J., Li J., and Luo L., Random Forests-Based Feature Selection for Land-Use Classification Using LIDAR Data and Orthoimagery, International Society for Photogrammetry and Remote Sensing, vol. XXXIX-B7, pp. 203-208, 2012.

[20] Hladnik A., Image Compression and Face Recognition: Two Image Processing Applications of Principal Component Analysis, International Circular of Graphic Education and Research, vol. 6, pp. 56-61, 2013.

[21] Inoue A. and Kilian L., How Useful is Bagging in Forecasting Economic Time Series? A Case Study of US CPI Inflation, CEPR Discussion Paper no. 5304, Available at: 276 The International Arab Journal of Information Technology, Vol. 15, No. 2, March 2018 http://ssrn.com/abstract=872856, Last Visited 2014.

[22] Jayakumar K., Revathi T., and Karpagam S., Intrusion Detection using Artificial Neural Networks with Best Set of Features, The International Arab Journal of Information Technology, vol. 12, no. 6A, pp. 728-734, 2015.

[23] Jolliffe I., Principal Component Analysis, John Wiley & Sons, 2014.

[24] Kavzoglu T. and Colkesen I., A Kernel Functions Analysis for Support: Vector Machines for Land Cover Classification, International Journal of Applied EarthObservation and Geoinformation, vol. 11, no. 5, pp. 352-359, 2009.

[25] Kim H. and Sohn G., Random Forests Based Multiple Classifier System for Power-Line Science Classification, in Proceedings of International Archives Photogram Remote Sensing Spatial Information Science, Canada, pp. 253-258, 2011.

[26] Kulkarni S. and Kelkar V., Classification of Multispectral Satellite Images Using Ensemble Techniques of Bagging, Boosting and Ada- Boost, in Proceedings of International Conference on Communication Circuits, Systems, and Information Technology Applications, Mumbai, pp. 253-258, 2014.

[27] Landsat-8., Using the USGS Landsat 8 Product, Available at: http://landsat.usgs.gov/Landsat8_Using_Product. php, Last Visited 2014.

[28] Maskova Z., Zemek F., and Kvet J., Normalized Difference Vegetation Index in the Management of Mountain Meadows, Boreal Environment Research, vol. 13, pp. 417-432, 2008.

[29] Mather P. and Tso B., Classification Methods for Remotely Sensed Data, CRC Press, 2009.

[30] McFeeters S., The Use of the Normalized Difference Water Index In the Delineation of Open Water Features, International Journal of Remote Sensing, vol. 17, no. 7, pp. 1425-1432, 1996.

[31] McFeeters S., Using the Normalized Difference Water Index within a Geographic Information System to Detect Swimming Pools for Mosquito Abatement: A Practical Approach, Remote Sensing, vol. 5, no. 7, pp. 3544-3561, 2013.

[32] Rouse J., Haas R., Schell J., and Deering D., Monitoring Vegetation Systems in the Great Plains with ERTS, NASA Goddard Space Flight Center, 3rd ERTS-1 Symposium, NASA, pp. 309- 317, 1974.

[33] Rumelhart D., Geoffrey E., and Robert J., Learning Internal Representations by Error Propagation, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Cambridge, pp. 318-362, 1986.

[34] Salah M., Trinder J., Shaker A., Hamed M., and Elsagheer A., Integrating Multiple Classifiers with Fuzzy Majority Voting for Improved Land Cover Classification, in Proceedings of WG III/2 Point Cloud ISPRS, At Saint-Mand , pp. 7- 12, 2010.

[35] Shlens J., A Tutorial on Principal Component Analysis, Cornell University Library Computer Science Learning, Available at: http://arxiv.org/abs/1404.1100, Last Visited 2015.

[36] Shota M. and Takuhiko M., Accuracy Comparison of Land Cover Mapping using The Object- Oriented Image Classification with Machine Learning Algorithms, in Proceedings of the 33rd Asian Conference on Remote Sensing, 2012.

[37] Tzeng Y., Fan K., and Chen K., An Adaptive Thresholding Multiple Classifiers System for Remote Sensing Image Classification, Photogrammetry Engineering and Remote Sensing, vol. 75, no. 6, pp. 679-687, 2009.

[38] Vapnik V. and Cortes C., Support-Vector Networks, Machine Learning, vol. 20, no. 3, pp. 273-297, 1995.

[39] Yang X., Deb S., and Fong S., Accelerated Particle Swarm Optimization and Support Vector Machine for Business Optimization and Applications, in Proceedings of 3rd International Conference Networked Digital Technologies Communications in Computer and Information Science, Macau, pp. 53-66, 2011.

[40] Zhuo Z., Boosting and Bagging of Neural Networks with Applications to Financial Time Series, Working paper, Department of Statistics, University of Chicago, 2006. Assessment of Ensemble Classifiers Using the Bagging Technique for ... 277 Hassan Mohamed was born in 1984 at Cairo, Egypt. He has got his Bachelor of Science degree in civil engineering (geomatics oriented), Faculty of Engineering at Shoubra, Benha University, Egypt in 2006. Hassan s master degree of science was in remote sensing and GIS from Geomatics Dept., Faculty of Engineering at Shoubra, Benha University, Egypt in 2012. Now, he is a PhD student at E-JUST. He worked as a demonstrator at the GeomaticsEngineering Department, Faculty of Engineering at Shoubra, Benha University, Egypt from 2007 to 2012. From 2012 till now, he has been an assistant lecturer at the Geomatics Engineering Department, Faculty of Engineering at Shoubra, Benha University, Egypt. Abdelazim Negm was born in Sharkia, Egypt. His background is civil engineering because he was graduated from Irrigation and Environmental Engineering Dept. in 1985. Prof. Negm has got his M.Sc. degree from Ain Shams University in 1990 in hydrology of the Nile basin. He got the PhD degree in 1992 inhydraulics. Currently, he is a professor of water resources in Egypt-Japan University for Science and Technology (E-JUST) since Oct. 2012 and the chairman of the Environmental Engineering Dept. at E-JUST since Feb. 17, 2013. His research areas are wide to include hydraulic, hydrology and water resources. He published about 200 papers in national and international journals and conferences. He participated in more than 55 conferences. He has awarded the prizes of best papers three times. He participates in the two EU funded international projects. For his detailed information one can visit his websites www.amneg.name.eg and www.amnegm.com. Mohamed Zahran is a professor in civil engineering (surveying and photogrammetry oriented). He was graduated from the Department of Geomatics Engineering Faculty of Engineering at Shoubra, Benha University in 1984. Prof. Zahran has got his M.Sc. degree from the department of Civil Engineering, Faculty of Engineering, Cairo University in 1989. He got the PhD degree from the Department of Civil and Geodetic Science, The Ohio State University in 1997. Currently, he is a professor of surveying and photogrammetry in Faculty of Engineering at Shoubra, Benha University since 2008 and a chairman of the Department of Geomatics Engineering Faculty of Engineering at Shoubra, Benha University since 2013. His research areas are wide to include digital photogrammetry, digital image analysis, remote sensing for mapping and close-range photogrammetry. He published many papers in national and international journals and conferenc Oliver Saavedra is a PhD in civil engineering (applied hydrology oriented). He is an associate professor at Tokyo Institute of Technology and adjunct professor to E-JUST since January 2010 to present. He has four years teaching experience in advanced hydrology, GIS, water resources tools for water resourcesmanagement lectures at graduate school. His major research interests are in development of decision supportingincluding optimal dam operation, flood control. He has about three years experience as a researcher (hydrology and WRM) and two years experience as a consultant engineer (water supply, sanitation, and infrastructure) and two years experience as a hydraulic engineer (water distribution systems). His project coordinator is Integrated water resources and environmental management for Asian and African mega-delta under climate change effects .