Identification of roofing materials with Discriminant Function Analysis and Random Forest classifiers on pan-sharpened WorldView-2 imagery – a comparison

  • Dávid Abriha Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary
  • Zoltán Kovács Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary
  • Sarawut Ninsawat Department of Remote Sensing and Geographic Information Systems (RS-GIS), Asian Institute of Technology, Bangkok, Thailand
  • László Bertalan Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary
  • Boglárka Balázs Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary
  • Szilárd Szabó Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary
Keywords: remote sensing, pan-sharpening, asbestos, machine learning


Identification of roofing material is an important issue in the urban environment due to hazardous and risky materials. We conducted an analysis with Discriminant Function Analysis (DFA) and Random Forest (RF) on WorldView-2 imagery. We applied a three- and a six-class approach (red tile, brown tile and asbestos; then dividing the data into shadowed and sunny roof parts). Furthermore, we applied pan-sharpening to the image. Our aim was to reveal the efficiency of the classifiers with a different number of classes and the efficiency of pan-sharpening. We found that all classifiers were efficient in roofing material identification with the classes involved, and the overall accuracy was above 85 per cent. The best results were gained by RF, both with three and with six classes; however, quadratic DFA was also successful in the classification of three classes. Usually, linear DFA performed the worst, but only relatively so, given that the result was 85 per cent. Asbestos was identified successfully with all classifiers. The results can be used by local authorities for roof mapping to build registers of buildings at risk.

Author Biographies

Dávid Abriha, Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary

Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary

Zoltán Kovács, Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary

Department of Physical Geography and Geoinformatics, University of Debrecen, Debrecen, Hungary


Abriha, D. 2017. Roofing material determination with hyperspectral data. Student V4 Geoscience Conference and Scientific Meeting GISÁČEK, Ostrava, Czech Republic, 5–10.

Allen, C., Tsou, M-H., Aslam, A., Nagel, A. and Gawron, J-M. 2016. Applying GIS and Machine Learning Methods to Twitter Data for Multiscale Surveillance of Influenza. PLoS ONE 11. (7): e0157734.

Alparone, L., Baronti, S., Garzelli, A. and Nencini, F. 2004. A global quality measurement of pansharpened multispectral imagery. IEEE Geoscience and Remote Sensing Letters 1. (4): 313–317.

Balázs B., Bíró, T., Dyke, G., Singh, K. and Szabó, Sz. 2018. Extracting water-related features using reflectance data and principal component analysis of Landsat images. Hydrological Sciences Journal 63. (2): 269–284.

Bandos, T.V., Bruzzone, L. and Camps-Valls, G. 2009. Classification of hyperspectral images with regularized linear discriminant analysis. IEEE Transactions on Geoscience and Remote Sensing 47. (3): 862–873.

Barakat, M.A., Shafri, H.Z.M. and Hamedianfar, A. 2017. New semi-automated mapping of asbestos cement roofs using rule-based object-based image analysis and Taguchi optimization technique from WorldView-2 images. International Journal of Remote Sensing 38. (2): 467–491.

Breiman, L. 2001. Random Forests. Machine Learning 45. (1): 5–32.

Brownlee, J. 2016. Machine Learning Mastery with R. Get Started, Build Accurate Models and Work Through Projects Step-by-Step. First edition, Machine Learning Mastery.

Burai, P., Lénárt, Cs., Valkó, O., Bekő, L., Szabó, Zs. and Deák, B. 2016. Fátlan vegetációtípusok azonosítása légi hiperspektrális távérzékelési módszerrel (Vegetation mapping in an alkali landscape – application of airborne hyperspectral data). Tájökológiai Lapok 14. (1): 1–12. (In Hungarian)

Burai, P., Tomor, T., Bekő, L. and Deák, B. 2015. Airborne hyperspectral remote sensing for identification grassland vegetation. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 40. (3): 427–431.

Chhikara, R.S. and Odell, P.L. 1973. Discriminant analysis using certain normed exponential densities with emphasis on remote sensing application. Pattern Recognition 5. (3): 259–272.

Cilia, C., Panigada, C., Rossini, M., Candiani, G., Pepe, M. and Colombo, R. 2015. Mapping of Asbestos Cement Roofs and Their Weathering Status Using Hyperspectral Aerial Images. ISPRS International Journal of Geo-Information 4. (2): 928–941.

Comber, A.J., Fisher, P., Brunsdon, C. and Khmag, A. 2012. Spatial analysis of remote sensing image classification accuracy. Remote Sensing of Environment 127. 237–246.

Congalton, R.G. 1991. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment 37. 35–46.

Deák, M., Telbisz, T., Árvai, M., Mari, L., Horváth, F., Kohán, B., Szabó, O. and Kovács, J. 2017. Heterogeneous forest classification by creating mixed vegetation classes using EO-1 Hyperion. International Journal of Remote Sensing 38. (18): 5215–5231.

Du, Q. and Nekovei, R. 2005. Implementation of real-time constrained linear discriminant analysis to remote sensing image classification. Pattern Recognition 38. (4): 459–471.

Du, Q. and Younan, N.H. 2008. Dimensionality Reduction and Linear Discriminant Analysis for Hyperspectral Image Classification. In Knowledge-Based Intelligent Information and Engineering Systems. Eds.: Lovrek, I., Howlett, R.J. and Jain, L.C., KES 2008. Lecture Notes in Computer Science 5179. 392–399.

Ehlers, M., Klonus, S., Åstrand, P.J. and Rosso, P. 2010. Multi-sensor image fusion for pansharpening in remote sensing. International Journal of Image and Data Fusion 1. (1): 25–45.

Environmental Systems Research Institute, National Center for Geographic Information and Analysis, and The Nature Conservancy 1994. Accuracy Assessment Procedures: NBS/NPS Vegetation Mapping Program. Santa Barbara, CA, and Arlington, VA. Report prepared for the National Biological Survey and National Park Service, Redlands.

Enyedi, P., Pap, M., Kovács, Z., Takács-Szilágyi, L. and Szabó, S. 2018. Efficiency of local minima and GLM techniques in sinkhole extraction from a LiDAR-based terrain model. International Journal of Digital Earth

EüM-KöM 2000. EüM-KöM decree on the limitations of the activities with dangerous materials and products. Magyar Közlöny 41. 125. (In Hungarian)

Fernández-Delgado, M., Cernadas, E., Barro, S. and Amorim, D. 2014. Do we need hundreds of classifiers to solve real world classification problems? Journal of Machine Learning Research 15. 3133–3181.

Gibril, M.B.A., Shafri, H.Z.M. and Hamedianfar, A. 2016. New semi-automated mapping of asbestos cement roofs using rule-based object-based image analysis and Taguchi optimization technique from WorldView-2 images. International Journal of Remote Sensing 38. (2): 467–491.

Gulácsi, A. and Kovács, F. 2018. Drought monitoring of forest vegetation using MODIS-based normalized difference drought index in Hungary. Hungarian Geographical Bulletin 67. (1): 29–42.

Hallouche, F., Adams, A.E., Hinton, O.R., Surtees, D.P., Wadehra, V. and Sherbet, G.V. 1993. Discriminant analysis for classification of murine melanomas and human cervical epithelial cells. Analytical and Quantative Cytollogy and Histology 15. 50–60.

Hijmans, R.J. 2017. raster: Geographic Data Analysis and Modeling. R package version 2.6-7.

Ho, T.K. 1995. Random Decision Forests. Proceedings of the 3rd International Conference on Document Analysis and Recognition, Montreal, QC, 14–16. August 1995. 278–282.

Kang, D., Myung, M-S., Kim, Y-K. and Kim, J-E. 2013. Systematic Review of the Effects of Asbestos Exposure on the Risk of Cancer between Children and Adults. Annals of Occupational and EnvironmentalMedicine 25(1) 10. Doi: 10.1186/2052-4374-25-10.

Kassambara, A. 2018. Machine Learning Essentials: Practical Guide in R. 1st edition. CreateSpace Independent Publishing Platform.

Krówczyńska, M., Wilk, E., Pabjanek, P., Zagajewski, B. and Meuleman, K. 2016. Mapping asbestoscement roofing with the use of APEX hyperspectral airborne imagery: Karpacz area, Poland – a case study. Miscellanea Geographica 20. (1): 41–46.

Książek, J. 2014. Methods for Detection of Asbestos-Cement Roofing Sheets. Geomatics and Environmental Engineering 8. (3): 59–76.

Kuhn, M., Wing, J., Weston, S., Williams, A., Keefer, C., Engelhardt, A., Cooper, T., Mayer, Z., Kenkel, B., R Core Team, Benesty, M., Lescarbeau, R., Ziem, A., Scrucca, L., Tang, Y., Canadan, C. and Hunt, T. 2018. caret: Classification and Regression Training. R package version 6.0–79.

Louppe, G., Wehenkel, L., Sutera, A. and Geurts, P. 2013. Understanding variable importances in forests of randomized trees. Proceedings of the 26th International Conference on Neural Information Processing Systems 1. 431–439.

Mándi, A., Posgay, M., Vadász, P., Major, K., Rödelsperger, K., Tossavainen, A., Ungváry, G., Woitowitz, H.J., Galambos, E., Németh, L., Soltész, I., Egerváry, M. and Böszörményi Nagy, G. 2000. Role of occupational asbestos exposure in Hungarian lung cancer patients. International Archives of Occupational and Environmental Health 73. 555–560.

Manickavasagan, A., Jayas, D.S. and White, N.D.G. 2008. Thermal imaging to detect infestation by Cryptolestes ferrugineus inside wheat kernels. Journal of Stored Products Research 44. (2): 186–192.

Martínez, A.M. and Kak, A.C. 2001. PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence 23. (2): 228–233.

Maurer, T. 2013. How to Pan-Sharpen Images Using the Gram-Schmidt Pan-Sharpen Method – a Recipe. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XL-1/W1. 239–244.

Mucsi, L., Liska, Cs.M., Henits, L., Tobak, Z., Csendes, B. and Nagy, L. 2017. The evaluation and application of an urban land cover map with image data fusion and laboratory measurements. Hungarian Geographical Bulletin 66. (2): 145–156.

Nagyváradi, L., Gyenizse, P. and Szebényi, A. 2011. Monitoring the changes of a suburban settlement by remote sensing. Acta Geographica Debrecina Landscape and Environment 5. 76–83.

Padwick, C., Deskevich, M., Pacifici, F. and Smallwood, S. 2010. WorldView-2 Pan-Sharpening. Proceedings of ASPRS 2010 Annual Conference, San Diego, CA, USA, 26–30.

Pal, M. 2005. Random forest classifier for remote sensing classification. International Journal of Remote Sensing 26. 217–222.

Pásztor, L., Laborczi, A., Takács, K., Szatmári, G., Dobos, E., Illés, G., Bakacsi, Zs. and Szabó, J. 2015. Compilation of novel and renewed, goal oriented digital soil maps using geostatistical and data mining tools. Hungarian Geographical Bulletin 64. (1): 49–64.

Petja, P.M., Twumasi, Y.A., Tengbeh, G.T. and Atanasova, M. 2010. Spatial epidemiology risk assessment for rehabilitated former asbestos mining areas in Limpopo Province, South Africa, using remote sensing and conventional analytical methods. South African Journal of Epidemiology and Infection 25. 32–39.

Podani, J. 2000. Introduction to the Exploration of Multivariate Biological Data. Leiden, Backhuys Publisher.

Powers, D.M.W. 2007. Evaluation: From Precision, Recall and F-Factor to ROC, Informedness, Markedness & Correlation. Technical Report SIE-07-001. Adelaide, South Australia. School of Informatics and Engineering, Flinders University Adelaide.

Pranckevičius, T. and Marcinkevičius, V. 2017. Comparison of Naive Bayes, Random Forest, Decision Tree, Support Vector Machines, and Logistic Regression Classifiers for Text ReviewsClassification. Baltic Journal of Modern Computing 5. (2): 221–232.

Raczko, E. and Zagajewski, B. 2017. Comparison of support vector machine, random forest and neural network classifiers for tree species classification on airborne hyperspectral APEX images. European Journal of Remote Sensing 50. (1): 144–154.

R Core Team 2018. R: A language and environment for statistical computing. Vienna, R Foundation for Statistical Computing.

Rouse, J.W. Jr., Haas, R.H., Deering, D.W., Schell, J.A. and Harlan, J.C. 1974. Monitoring the Vernal Advancement and Retrogradation (Green Wave Effect) of Natural Vegetation. NASA/GSFC Type III Final Report, Greenbelt, MD.

Samsudin, S.H., Shafri, H.Z.M. and Hamedianfar, A. 2016. Development of spectral indices for roofing material condition status detection using field spectroscopy and WorldView-3 data. Journal of Applied Remote Sensing 10(2) 025021.

Siqueira, L.F.S., Araújo Júnior, R.F.A., de Araújo, A.A., Morais, C.L.M. and Lima, K.M.G. 2017. LDA vs. QDA for FT-MIR prostate cancer tissue classification. Chemometrics and Intelligent Laboratory Systems 162. 123–129.

Statnikov, A., Wang, L. and Aliferis, C. 2008. A comprehensive comparison of random forests and support vector machines for microarray-based cancer classification. BMC Bioinformatics 9. 319.

Stevens, J. 1996. Applied multivariate statistics for the social sciences. 3rd edition. Mahwah, NJ, USA, Lawrence Erlbaum.

Switzer, P. 1980. Extensions of linear discriminant analysis for statistical classification of remotely sensed satellite imagery. Journal of the International Association for Mathematical Geology 12. (4): 367–376.

Szabó, S., Gácsi, Z. and Balázs, B. 2016. Specific features of NDVI, NDWI and MNDWI as reflected in land cover categories. Acta Geographica Debrecina Landscape and Environment 10. (3–4): 194–202.

Szabó, Sz., Burai, P., Kovács, Z., Szabó, Gy., Kerényi, A., Fazekas, I., Paládi, M., Buday, T. and Szabó, G. 2014. Testing of algorithms for the identification of asbestos roofing based on hyperspectral data. Environmental Engineering and Management Journal 143. (11): 2875–2880.

Szabó, Z., Tóth, C.A., Tomor, T. and Szabó, S. 2017. Airborne LiDAR point cloud in mapping of fluvial forms: a case study of a Hungarian floodplain. GIScience and Remote Sensing 54. 862–880.

Taherzadeh, E. and Shafri, H.Z.M. 2013. Development of a Generic Model for the Detection of Roof Materials Based on an Object-Based Approach Using WorldView-2 Satellite Imagery. Advances in Remote Sensing 2. (4): 312–321.

Taherzadeh, E., Shafri, H.Z.M. and Shahi, K. 2014. Roof Material Detection Based on Object-Based Approach Using WorldView-2 Satellite Imagery. International Journal of Environmental and Ecological Engineering 8. (10): 1826–1829.

Tharwat, A. 2016. Linear vs. quadratic discriminant analysis classifier: a tutorial. International Journal of Applied Pattern Recognition 3. (2): 145–180.

Therneau T. and Atkinson, B. 2018. rpart: Recursive Partitioning and Regression Trees. R package version 4.1–13.

Vadivambal, R., Vellaichamy, C., Jayas, D.S. and White, N.D.G. 2010. Detection of Sprout-Damaged Wheat Using Thermal Imaging. Applied Engineering in Agriculture 26. 999–1004.

Venables, W.N. and Ripley, B.D. 2002. Modern Applied Statistics with S. Fourth edition. New York, Springer.

Wickham, H. 2017. Tidyverse: Easily Install and Load the 'Tidyverse'. R package version 1.2.1.

Wilk, E., Krówczyńska, M. and Pabjanek, P. 2015. Determinants influencing the amount of asbestoscement roofing in Poland. Miscellanea Geographica 19. (3): 82–86.

Wina, Herwindiati, D.E. and Isa, S.M. 2014. Robust discriminant analysis for classification of remote sensing data. International Conference on Advanced Computer Science and Information System, Jakarta, 18–19. Oct. 2014. Jakarta, IEEE Indonesia Section, 454–458.

Yuhendra, Alimuddin, I., Sumantyo, J.T.S. and Kuze, H. 2012. Assessment of pan-sharpening methods applied to image fusion of remotely sensed multi-band data. International Journal of Applied Earth Observation and Geoinformation 18. 165–174.

How to Cite
Abriha, D., Kovács, Z., Ninsawat, S., Bertalan, L., Balázs, B., & Szabó, S. (2018). Identification of roofing materials with Discriminant Function Analysis and Random Forest classifiers on pan-sharpened WorldView-2 imagery – a comparison. Hungarian Geographical Bulletin, 67(4), 375-392.