{"title":"Fusion of ETM+ Multispectral and Panchromatic Texture for Remote Sensing Classification","authors":"Mahesh Pal","volume":15,"journal":"International Journal of Electrical and Computer Engineering","pagesStart":413,"pagesEnd":416,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/11228","abstract":"This paper proposes to use ETM+ multispectral data\r\nand panchromatic band as well as texture features derived from the\r\npanchromatic band for land cover classification. Four texture features\r\nincluding one 'internal texture' and three GLCM based textures\r\nnamely correlation, entropy, and inverse different moment were used\r\nin combination with ETM+ multispectral data. Two data sets\r\ninvolving combination of multispectral, panchromatic band and its\r\ntexture were used and results were compared with those obtained by\r\nusing multispectral data alone. A decision tree classifier with and\r\nwithout boosting were used to classify different datasets. Results\r\nfrom this study suggest that the dataset consisting of panchromatic\r\nband, four of its texture features and multispectral data was able to\r\nincrease the classification accuracy by about 2%. In comparison, a\r\nboosted decision tree was able to increase the classification accuracy\r\nby about 3% with the same dataset.","references":"[1] Lu, D and Weng, Q, Urban Classification Using Full Spectral\r\nInformation of Landsat ETM+ Imagery in Marion County, Indiana,\r\nPhotogrammetric Engineering & Remote Sensing, 71(11), 2005, 1275-\r\n1284.\r\n[2] L.,Breiman, J.H. Friedman, R.A.,Olshen and C.J.,Stone, Classification\r\nand Regression Trees, Wadsworth, Monterey, CA, 1984.\r\n[3] S. K.Murthy, S. Kasif and S. Salzberg, A system for induction of oblique\r\ndecision trees. Journal of Artificial Intelligence Research, 2, 1994, 1-32.\r\n[4] I. Kononenko and J. S. Hong, Attribute selection for modelling. Future\r\nGeneration Computer Systems, 13, 1997, 181-195.\r\n[5] J. Mingers, An empirical comparison of selection measures for decision\r\ntree induction. Machine Learning, 3, 1989, 319-342.\r\n[6] J. R. Quinlan, C4.5: Programs for Machine Learning. San Mateo:\r\nMorgan Kaufmann, San Francisco, 1993.\r\n[7] Y.Freund and R. Schapire, Experiments with a new boosting algorithm.\r\nMachine Learning: Proceedings of the Thirteenth International\r\nconference, 148-156, 1996.\r\n[8] M. A. Friedl, C. E. Brodley, and A. H. Strahler, Maximizing land cover\r\nclassification accuracies produced by decision tree at continental to\r\nglobal scales. IEEE Transactions on Geoscience and Remote Sensing.\r\n37, 1999, 969-977.\r\n[9] M. Pal and P. M. Mather, An assessment of the effectiveness of decision\r\ntree methods for land cover classification. Remote Sensing of\r\nEnvironment. 86, 2003, 554-565.\r\n[10] G. J. Briem, J. A. Benediktsson, and J. R. Sveinsson, Multiple\r\nClassifiers Applied to Multisource Remote Sensing Data IEEE\r\nTransactions on Geoscience and Remote Sensing, 40, 2002, 2291-2299.\r\n[11] R.M. Harlick, K. Shanmugam and I. Dinstein, Texture features for image\r\nclassification. IEEE Transactions on System, Man, and Cybernetics,\r\n3(6), 1973, 610-621.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 15, 2008"}