A comparison of the classification accuracies in determining the land cover of Kadirli region of Turkey by using the pixel based and object based classification algorithms

Year 2017, , 52 - 60, 01.06.2017

Pınar Karakuş , Hakan Karabork Sinasi Kaya

https://doi.org/10.26833/ijeg.298951

Cited By: 17

Abstract

Pixel and object-based classification methods have been used for the determination of land cover. Pixel based classification methods suffer from salt and pepper effect. So pixel based classification methods cannot reach the accuracy of the object based classification. In order to eliminate the salt and pepper effect on the remote sensing classification accuracy and improve the result maps created as a result of the classification and further improve the classification accuracy in pixel based classification, it is recommended that the sieve class, clump class and majority analyses -which are ordinarily applied to high resolution images in this study by using the pixel based classification method. So the effect of these analyzes on low and medium resolution satellite images are unknown. With the SPOT 5 satellite image, this study will investigate how much this analysis affects the accuracy of classification. The classification includes the following categories: sun flowers, corns, peanuts, trees, roads, residential areas and water resources. In this study, the object based classification method was compared with three pixel based classification methods, namely the support vector machines, maximum likelihood method and spectral angle mapper method. The following general accuracy and kappa values were obtained from the methods in question: Object based classification method (96% accuracy, kappa value of 0,949), maximum likelihood method (90.99% accuracy, kappa value of 0,67), support vector machines (92.06 accuracy, kappa value of 0.70), spectral angle mapper method (93.88% accuracy, kappa value of 0,78). Following the pixel based classification process, the total accuracy and kappa values of the classified image was improved through the application of sieve class, clump class and majority analyses. As a result of the analyses conducted on the pixel based classification methods, the following general accuracy and kappa values were obtained for the following pixel based classification methods: maximum likelihood method (92.91% accuracy, kappa value of 0,73), support vector machines (93.13% accuracy, kappa value of 0.74) and spectral angle mapper method (95.62% accuracy, kappa value of 0,88). As a result of the analyses applied to the pixel based classification method, the classification accuracy produced similar results to that of the object based classification accuracy. To the best knowledge our author this is the first study dealing with this study area. So the authors think that this paper present a different point of view for interested researchers in this study area.

Keywords

Object based classification, maximum likelihood method, support vector machine, spectral angle mapper, post classification

References

• Baatz, M., Schape, A., 2000. Multi resolution segmentation-an optimization approach for high quality multi scale image segmentation. Angewandte Geographische Informations verarbeitung XII. Karlsruhe, Germany, pp. 12−23.
• Blaschke, T. 2010. Object Based Image Analysis for Remote Sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 65, 2-16.
• Buddenbaum, H., Schlerf M.,Hill J., 2005. Classification of coniferous tree species and age classes using hyperspectral data and geostatistical methods. International Journal of Remote Sensing, 26( 24),5453- 5465.
• Burges, C.,J,C., 1998. ATutorial on Support Vector Machines for Pattern Recognition. Data Mining and Knowledge Discovery, 2,121-167.
• Casals-Carrasco, P., Kubo,S., Babu Madhavan, B., 2000. Application of Spektral Mixture Analysis For Terrain Evaluation Studies. International Journal Of Remote Sensing, 21, 3039-3055.
• Castillejo-Gonzalez I.L., López-Granados F., Garcia- Ferrer A., Peña-Barragan J.M., Jurado-Expósito Sanchez-de la Orden M., Gonzalez-Audicana M., 2009. Object- and pixel-based analysis for mapping crops and their agro-environmental associated measures using QuickBird imagery. Computers and Electronics in Agriculture, 68,207–215
• Castro, A. I., Jurado-Expósito, M., Pena-Barragan, J. M., López-Granados F., 2012. Airborne multi-spectral imagery for mapping cruciferous weeds in cereal and legume crops. Precision Agriculture, 13 (3), 302-321.
• Congalton, R.G., 1991. A review of assessing the accuracy of classification of remotely sensed data.Remote Sens. Environ., 37 ,35–46.
• Definiens, Developer 7, User Guide
• Dey, V., 2011. A Supervised Approach for the Estimation of Parameters of Multiresolution Segmentation and its Application in Building Feature Extraction from VHR Imagery. M.Sc.E Thesis, Department of Geodesy and Geomatics Engineering Technical Report No. 278, University of New Brunswick, Fredericton, New Brunswick, Canada, 162.
• Dragut, L., Tiede, D., Levick, S.R., 2010. ESP: a tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data, International Journal of Geographical Information Science, 24, 859– 871
• Dymond, J. R., Shepherd, J. D. 2004. The spatial distribution of indigenous forest and its composition in the Wellington region, New Zealand, from ETM+ satellite imagery. Remote Sensing of Environment, 90(1),116–125.
• Ekercin, S., 2007. Uzaktan Algılama Ve Coğrafi Bilgi Sistemleri Entegrasyonu İle Tuz Gölü Ve Yakın Çevresinin Zamana Bağlı Değişim Analizi. Doktora Tezi, İTÜ.
• Faria, F.,A.,Dos Santos, J.,A.,Torres, R. Da S., Rocha, A., Falcao, A., 2012. Automatic Fusion Of Region- Based Classifiers For Coffee Crop Recognition. IEEE, 978-1-4673-1159-5/12.
• Foody, G.M., 2002. Status of land cover classification accuracy assessment,Remote Sens. Environ., 80 , 185– 201
• Foody, G.M., Mathur, A., 2004. A Relative Evaluation Of Multiclass Image Classification By Support Vector Machines. IEEE Transactions On Geoscience And Remote Sensing, 42, 1335–1343.
• Gualtieri, J.A., Cromp, R.F, 1998. Support Vector Machines For Hyperspectral Remote Sensing Classification. Proc. SPIE, 3584, 221–232.
• Gürsoy, Ö., Kaya, Ş., Çakır, Z., 2013. Uydu görüntüleri ile yersel spektral ölçme verilerinin entegrasyonu. Havacılık ve Uzay Teknolojileri Dergisi,6(1),45-51.
• Hoberg, T., Müller, S., 2011. Multitemporal Crop Type Classification Using Conditional Random Fields and RapidEye Data. ISPRS, XXXVIII-4/W19, 115-121.
• Hong, G., Zhang,Y, Zhang, A.,Zhou, F, Li, J., 2007. Fusion of Modis And Radarsat Data For Crop Type Classification- An Initial Study. ISPRS Workshop On Updating Geo-Spatial Databases With Imagery & The 5th ISPRS Workshop On Dmgıss,Urumqin Xinjiang, Uygur,China.
• Huang, C., Davis, L.S., Townshend, J.R.G., 2002. An Assesment of support vector Machines for Land Cover Classification. Int. J. Remote Sensing, 23,725-749.
• İTÜ-UHUZAM Remote Sensing Laboratory www.cscrs.itu.edu.tr Accessed 19.10.2016
• Jensen, J. R., Garcia-Quijano, M., Hadley, B., Im, J., Wang, Z., Nel, A. L.,Teixeira, E., Davis, B. A., 2006. Remote Sensing Agricultural Crop Type For Sustainable Development In South Africa. Geocarto International, 21( 2), 5-18.
• Kruse,F.A.,Boardman, J.W., and Huntington, J.F., 2003. Comparison of airborne hyperspectral data and EO-1 Hyperion for mineral mapping.IEEE Transactionson Geoscience and Remote Sensing, 41,1388-1400.
• Kumar, P.,Dileep Gupt,a K., Mishra, V. N.,Prasad R., 2015. Comparison of support vector machine, artificial neural network, and spectral angle mapper algorithms for crop classification using LISS IV Data. International Journal of Remote Sensing, 36( 6), 1604–1617.
• Landis, J.R., Kock, G.G., 1977. The measurement of observer agreement for categorical data Biometrics, 33 ,159–174
• Lunetta, R.S.; Balogh, M.E., 1999. Application of multitemporal Landsat 5 TM imagery for wetland Identification. Photogramm. Eng. Remote Sens. 65, 1303–1310.
• Maxwell, S.K.; Nuckols, J.R.; Ward, M.H.; Hoffer, R.M., 2004. An automated approach to mapping corn from Landsat imagery. Comput. Electron. Agric., 43, 43–54.
• Melgani, F., Bruzzone, L., 2004. Classification of Hyperspectral Remote Sensing Images With Support Vector Machines, IEEE Transactions On Geoscience And Remote Sensing, 42(8), 1778–1790.
• Meneguzzo, D.M. , Liknes, G. C. , Nelson, M. D., 2013. Mapping trees outside forests using high-resolution aerial imagery: a comparison of pixel- and object-based classification approaches. Environmental Monitoring and Assessment. 185(8), 6261–6275.
• Michez, A. , Piégay, H. , Lisein, J. , Claessens, H., Lejeune, P., 2016. Classification of riparian forest species and health condition using multi-temporal and hyperspatial imagery from unmanned aerial system. Environmental Monitoring and Assessment.188:146.
• Montserud R.A., Leamans R., 1992. Comparing global vegetation maps with the kappa statistic Ecol. Model., 62 , 275–293.
• Murai, H.; Omatu, S., 1997. Remote sensing image analysis using a neural network and knowledge-based processing. Int. J. Remote Sens.,18, 811–828.
• Myint,S.W., Gober, P.,Brazel, A., Grossman-Clarke,S., Weng,Q., 2011. Per-pixel vs. object-based classification of urban land cover extraction using high spatial resolution imagery. Remote sensing of environment, 115(5),1145-1161.
• Naguib, A.M., Farag, M.A., Yahia, M.A., Ramadan H.H., Abd Elwahab M.S., 2009. Comparative study between support vector machines and neural networks for lithological discrimination using hyperspectral data. Egypt Journal of Remote Sensing and Space Science, 12,27–42
• Oommen, T., Misra, D., Twarakavi, N.K.C., Prakash A., Sahoo B., Bandopadhyay S., 2008. An objective analysis of support vector machine based classification for remote sensing. Mathematical Geosciences, 40, 409–422
• Orhan, O., Ekercin, S., Dadaser-Celik, F., 2014.Use of Landsat Land Surface Temperature and Vegetation Indices for Monitoring Drought in the Salt Lake Basin Area, Turkey. The Scientific World Journal, doi:10.1155/2014/142939
• Pal, M., Mather, P.M., 2005. Support Vector Machines For Classification in Remote Sensing. International Journal Of Remote Sensing, 26,1007-1011.
• Pena-Barragan, J.,M., Ngugi, M.,K.,Plant, R.,E., Six, J., 2011. Object-Based Crop Identification using Multiple Vegetation Indices, Textural Features and Crop Phenology. Remote Sensing Of Environment, 115,1301- 1306.
• Platt, R.V, Rapoza, L., 2008. An evaluation of an object- oriented paradigm for land use/land cover classification. the Professional geographer, 60(1),87
• Richards, J.A., Jia, X., 2006. Remote Sensing Digital Image Analysis(4. Edition), Germany:Springer.
• Rogan, J., Franklin, J., Roberts, D.A., 2002. A comparison of methods for monitoring multitemporal vegetation change using Thematic Mapper imagery. Remote Sens. Environ., 80 ,143–156.
• Sakieh, Y. , Gholipour, M. , Salmanmahiny, A., 2016. An integrated spectral-textural approach for environmental change monitoring and assessment: analyzing the dynamics of green covers in a highly developing region. Environmental Monitoring and Assessment.188:205.
• Sun, L., and Schulz, K., 2015. The Improvement of Land Cover Classification by Thermal Remote Sensing. remote sensing, 7, 8369-8391.
• Sunar, F.,Özkan, Ç., Osmanoğlu, B., 2016. Uzaktan algılama, T.C. Anadolu Üniversitesi Yayını, No:2320, 4. Edition.
• Vapnik, V.N., 1995. The Nature Of Statistical Learning Theory (1. Edition),New York: Springer-Verlag.
• Vapnik, V.N., 2000. The Nature Of Statistical Learning Theory ( 2. Edition) New York: Springer-Verlag.
• Varela, R. A. D., Ramil, Rego P. , Calvo, Iglesias S. , Muñoz Sobrino, C., 2008. Automatic Habitat Classification Methods Based On Satellite Images: A Practical Assessment In The Nw Iberia Coastal Mountains. Environmental Monitoring And Assessment.144(1),229-250.
• Weih, R. C., Jr. and Riggan, N. D., Jr., 2010. Object- Based Classıfıcatıon Vs. Pıxel-Based Classıfıcatıon: Comparıtıve Importance Of Multı-Resolutıon Imagery. The International Archives Of The Photogrammetry, Remote Sensing And Spatial Information Sciences, XXXVIII-4/C7.
• Whiteside, T. G., Boggs, G. S., Maier, S. W, 2011. Comparing object-based and pixel-based classifications for mapping savannas. International Journal of Applied Earth Observation and Geoinformation, 13, 884-893.
• Willhauck, G., Schneider, T., De Kok, R., Ammer U., 2000. Comparison of object-oriented classification techniques and standard image analysis for the use of change detection between SPOT multispectral satellite images and aerial photos.Proceedings of XIX ISPRS Congress, Amsterdam, July 16–22.
• Yan,G.,Mas,J.F.,Maathuis, B.H.P.,Xiangmin, Z., Van Dijk, P.M, 2006. Comparison of pixel based and object oriented iamge classification approaches-A case study in a coal fire area, Wuda, İnner Mongolia, China, International Journal of Remote Sensing, 27,4039-4055.
• Yang, C., Everitt, J. H., Murden, D., 2011. Evaluating high resolution SPOT 5 satellite imagery for crop identification .Computers and Electronics in Agriculture 75( 2), 347–354.
• Yonezawa, C., 2007. Maximum likelihood classification combined with spectral angle mapper algorithm for high resolution satellite imagery.International journal of Remote Sensing, 28,3729-3737.
• Zhang, Y.J., 1997. Evaluation And Comparison of Different Segmentation Algorithms. Pattern Recogniton Letters, 18,963-974.
• URL-1 http://www.harrisgeospatial.com/docs/SievingClasses.ht ml Accessed 19.10.2016.