The Potential of UAV Derived Image Features for Discriminating Savannah Tree Species



Mapping tree species at the single-tree level is an active field of research linking ecology and remote sensing. However, the discrimination of tree species requires the selection of the relevant spectral features derived from imagery. We can extract an extensive number of image parameters even from images with a low spectral resolution, such as Red-Green-Blue (RGB) or near-infrared (NIR) images. Hence, identifying the most relevant image parameters for tree species discrimination is still an issue. We generated 42 parameters from very high resolution images acquired by Unmanned Aerial Vehicles (UAV), such as chromatic coordinates, spectral indices, texture measures and a canopy height model (CHM). The aim of this study was to compare the relevance of these components for classifying savannah tree species. We obtained very high (5 cm) pixel resolution RGB-NIR imagery with a delta-wing UAV in a thorn bush savannah landscape in central Namibia in April 2016. Simultaneously, we gathered ground truth data on the location of 478 individual trees and large shrubs belonging to 16 species. We then used a Random Forest classifier on single and combined thematic sets of image data, e.g. RGB, NIR, texture and in combination with CHM. The best average overall accuracy was 0.77 and the best Cohen´s Kappa value was 0.63 for a combination of RGB imagery and the CHM. Our results are comparable to other studies using hyperspectral data and LiDAR information. We further found that the abundance of the tree species is crucial for successful mapping, with only species with a high abundance being classified satisfactorily. Diverse ecosystems such as savannahs could therefore be a challenge for future tree mapping projects. Nevertheless, this study indicates that UAV-borne RGB imagery seems promising for detailed mapping of tree species.


Biodiversity monitoring Drone GLCM LAS Namibia NDVI Point cloud Spectral discrimination 



Our gratitude to the Pommersche Farmereigesellschaft and their staff for allowing us to work on the farm Erichsfelde. The work was financially supported by the SASSCAL initiative, with funding by the German Federal Ministry of Education and Research; BMBF Funding Nr: 01LG1201M.


  1. Baldeck, C.A., Asner, G.P., Martin, R.E., et al.: Operational tree species mapping in a diverse tropical forest with airborne imaging spectroscopy. PLoS One. 10, e0118403 (2015). doi: 10.1371/journal.pone.0118403 CrossRefPubMedPubMedCentralGoogle Scholar
  2. Baret, F., Guyot, G., Major, D.: TSAVI: a vegetation index which minimizes soil brightness effects on LAI and APAR estimation. In: 12th Canadian Symposium on Remote Sensing and IGARSS’90, p. 4, Vancouver, Canada, 10–14 July 1989 (1989)Google Scholar
  3. Breiman, L.: Random forests. Mach. Learn. 45, 5–32 (2001). doi: 10.1023/A:1010933404324 CrossRefGoogle Scholar
  4. Bunting, P., Lucas, R.: The delineation of tree crowns in Australian mixed species forests using hyperspectral Compact Airborne Spectrographic Imager (CASI) data. Remote Sens. Environ. 101, 230–248 (2006). doi: 10.1016/j.rse.2005.12.015 CrossRefGoogle Scholar
  5. Cho, M.A., Mathieu, R., Asner, G.P., et al.: Mapping tree species composition in South African savannas using an integrated airborne spectral and LiDAR system. Remote Sens. Environ. 125, 214–226 (2012). doi: 10.1016/j.rse.2012.07.010 CrossRefGoogle Scholar
  6. Cho, M.A., Malahlela, O., Ramoelo, A.: Assessing the utility WorldView-2 imagery for tree species mapping in South African subtropical humid forest and the conservation implications: Dukuduku forest patch as case study. Int. J. Appl. Earth Obs. Geoinf. 38, 349–357 (2015). doi: 10.1016/j.jag.2015.01.015 CrossRefGoogle Scholar
  7. Colgan, M.S., Baldeck, C.A., Féret, J.-B., Asner, G.P.: Mapping savanna tree species at ecosystem scales using support vector machine classification and BRDF correction on airborne hyperspectral and LiDAR data. Remote Sens. 4, 3462–3480 (2012). doi: 10.3390/rs4113462 CrossRefGoogle Scholar
  8. Conrad, O., Bechtel, B., Bock, M., et al.: System for automated geoscientific analyses (SAGA) v. 2.1.4. Geosci. Model Dev. 8, 1991–2007 (2015). doi: 10.5194/gmd-8-1991-2015 CrossRefGoogle Scholar
  9. Culvenor, D.S.: TIDA: an algorithm for the delineation of tree crowns in high spatial resolution remotely sensed imagery. Comput. Geosci. 28, 33–44 (2002). doi: 10.1016/S0098-3004(00)00110-2 CrossRefGoogle Scholar
  10. Dalponte, M., Coomes, D.A.: Tree-centric mapping of forest carbon density from airborne laser scanning and hyperspectral data. Methods Ecol. Evol. 7, 1236–1245 (2016). doi: 10.1111/2041-210X.12575 CrossRefPubMedPubMedCentralGoogle Scholar
  11. Duro, D.C., Franklin, S.E., Dubé, M.G.: A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery. Remote Sens. Environ. 118, 259–272 (2012). doi: 10.1016/j.rse.2011.11.020 CrossRefGoogle Scholar
  12. Dvořák, P.J., Müllerová J., Bartaloš, T., Brůna J.: Unmanned aerial vehicles for alien plant species detection and monitoring. ISPRS – international archives of the photogrammetry, remote sensing and spatial information sciences XL-1/W4: 83–90 (2015). doi:10.5194/isprsarchives-XL-1-W4-83-2015Google Scholar
  13. Fassnacht, F.E., Latifi, H., Stereńczak, K., et al.: Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ. 186, 64–87 (2016). doi: 10.1016/j.rse.2016.08.013 CrossRefGoogle Scholar
  14. Fischer, T., Veste, M., Eisele, A., et al.: Small scale spatial heterogeneity of Normalized Difference Vegetation Indices (NDVIs) and hot spots of photosynthesis in biological soil crusts. Flora – Morphol. Distrib. Funct. Ecol. Plants. 207, 159–167 (2012). doi: 10.1016/j.flora.2012.01.001 CrossRefGoogle Scholar
  15. Giess, W.: A preliminary vegetation map of Namibia. Dinteria. 4, 1–112 (1998)Google Scholar
  16. Haralick, R.M., Shanmugam, K., Dinstein, I.: Textural features for image classification. IEEE Trans. Syst. Man Cybern. 3, 610–621 (1973)CrossRefGoogle Scholar
  17. Huete, A.R.: A soil-adjusted vegetation index (SAVI). Remote Sens. Environ. 25, 295–309 (1988). doi: 10.1016/0034-4257(88)90106-X CrossRefGoogle Scholar
  18. Immitzer, M., Atzberger, C., Koukal, T.: Tree species classification with random forest using very high spatial resolution 8-Band WorldView-2 satellite data. Remote Sens. 4, 2661–2693 (2012). doi: 10.3390/rs4092661 CrossRefGoogle Scholar
  19. Isenburg, M.: LAStools, efficient LiDAR processing software. rapidlasso GmbH (2016)Google Scholar
  20. Jürgens, N., Schmiedel, U., Haarmeyer, D.H., et al.: The BIOTA biodiversity observatories in Africa—a standardized framework for large-scale environmental monitoring. Environ. Monit. Assess. 184, 655–678 (2012). doi: 10.1007/s10661-011-1993-y CrossRefPubMedGoogle Scholar
  21. Kang, J., Wang, L., Chen, F., Niu, Z.: Identifying tree crown areas in undulating eucalyptus plantations using JSEG multi-scale segmentation and unmanned aerial vehicle near-infrared imagery. Int. J. Remote Sens. 38, 1–17 (2016). doi: 10.1080/01431161.2016.1253900 CrossRefGoogle Scholar
  22. Klaassen, E.S., Kwembeya, E.G.: A checklist of Namibian indigenous and naturalised plants. 2013. Occasional Contributions No 5, National Botanical Research Institute, Windhoek, Namibia (2013).Google Scholar
  23. Krefis, A.C., Schwarz, N.G., Nkrumah, B., et al.: Spatial analysis of land cover determinants of malaria incidence in the Ashanti Region, Ghana. PLoS One. 6, e17905 (2011). doi: 10.1371/journal.pone.0017905 CrossRefPubMedPubMedCentralGoogle Scholar
  24. Kuhn, M., Johnson, K.: Applied Predictive Modeling. Springer, New York (2013)CrossRefGoogle Scholar
  25. Kuhn, M.K., Weston, S., Williams, A., et al.: Caret: Classification and Regression Training. R package version 6.0-70. (2016)
  26. Kyalangalilwa, B., Boatwright, J.S., Daru, B.H., et al.: Phylogenetic position and revised classification of Acacia s.l. (Fabaceae: Mimosoideae) in Africa, including new combinations in Vachellia and Senegalia. Bot. J. Linn. Soc. 172, 500–523 (2013). doi: 10.1111/boj.12047 CrossRefGoogle Scholar
  27. Liaw, A., Wiener, M.: Classification and regression by random Forest. R News. 2, 18–22 (2002)Google Scholar
  28. Lisein, J., Michez, A., Claessens, H., Lejeune, P.: Discrimination of deciduous tree species from time series of unmanned aerial system imagery. PLoS One. 10, e0141006 (2015). doi: 10.1371/journal.pone.0141006 CrossRefPubMedPubMedCentralGoogle Scholar
  29. Lucas, R., Bunting, P., Paterson, M., Chisholm, L.: Classification of Australian forest communities using aerial photography, CASI and HyMap data. Remote Sens. Environ. 112, 2088–2103 (2008). doi: 10.1016/j.rse.2007.10.011 CrossRefGoogle Scholar
  30. Magurran, A.E., McGill, B.J.: Biological Diversity: Frontiers in Measurement and Assessment, 1st edn. Oxford University Press, Oxford/New York (2011).Google Scholar
  31. McInerney, D., Kempeneers, P.: Orfeo toolbox. In: Open Source Geospatial Tools, pp. 199–217. Springer International Publishing, Basel (2015).Google Scholar
  32. Meyer, G.E., Neto, J.C.: Verification of color vegetation indices for automated crop imaging applications. Comput. Electron. Agric. 63, 282–293 (2008). doi: 10.1016/j.compag.2008.03.009 CrossRefGoogle Scholar
  33. Meyer, D., Dimitriadou, E., Hornik, K., et al.: e1071: Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), TU Wien. R package version 1.6-7 (2015)
  34. Naidoo, L., Cho, M.A., Mathieu, R., Asner, G.: Classification of savanna tree species, in the Greater Kruger National Park region, by integrating hyperspectral and LiDAR data in a Random Forest data mining environment. ISPRS J. Photogramm. Remote Sens. 69, 167–179 (2012). doi: 10.1016/j.isprsjprs.2012.03.005 CrossRefGoogle Scholar
  35. Pal, M.: Random forest classifier for remote sensing classification. Int. J. Remote Sens. 26, 217–222 (2005). doi: 10.1080/01431160412331269698 CrossRefGoogle Scholar
  36. Perry, C.R., Lautenschlager, L.F.: Functional equivalence of spectral vegetation indices. Remote Sens. Environ. 14, 169–182 (1984). doi: 10.1016/0034-4257(84)90013-0 CrossRefGoogle Scholar
  37. QGIS Development team. QGIS Geographic Information System. Open Source Geospatial Foundation (2016)Google Scholar
  38. Qi, J., Chehbouni, A., Huete, A.R., Kerr, Y.H., Sorooshian, S.: A modified soil adjusted vegetation index. Remote Sens. Environ. 48, 119–126 (1994). doi: 10.1016/0034-4257(94)90134-1 CrossRefGoogle Scholar
  39. Qian, Y., Zhou, W., Yan, J., et al.: Comparing machine learning classifiers for object-based land cover classification using very high resolution imagery. Remote Sens. 7, 153–168 (2014). doi: 10.3390/rs70100153 CrossRefGoogle Scholar
  40. R Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna (2016)Google Scholar
  41. Rasmussen, J., Nielsen, J., Garcia-Ruiz, F., Christensen, S., Streibig, J.C.: Potential uses of small unmanned aircraft systems (UAS) in weed research. Weed Res. 53, 242–248 (2013). doi: 10.1111/wre.12026 CrossRefGoogle Scholar
  42. Rasmussen, J., Ntakos, G., Nielsen, J., et al.: Are vegetation indices derived from consumer-grade cameras mounted on UAVs sufficiently reliable for assessing experimental plots? Eur. J. Agron. 74, 75–92 (2016). doi: 10.1016/j.eja.2015.11.026 CrossRefGoogle Scholar
  43. Richards, J.A.: Remote Sensing Digital Image Analysis: An Introduction, 5th edn. Springer, Berlin (2013)CrossRefGoogle Scholar
  44. Schirrmann, M., Giebel, A., Gleiniger, F., et al.: Monitoring agronomic parameters of winter wheat crops with low-cost UAV imagery. Remote Sens. 8, 706 (2016). doi: 10.3390/rs8090706 CrossRefGoogle Scholar
  45. SenseFly. User Manual: S110 RGB/NIR /RE camera. SenseFly Ltd., Lausanne, Switzerland (2014)Google Scholar
  46. SenseFly. eBee Sensefly: Extended User MANUAL eBee and eBee Ag. Revision 17, June 2015. SenseFly Ltd., Lausanne, Switzerland (2015)Google Scholar
  47. Silleos, N.G., Alexandridis, T.K., Gitas, I.Z., Perakis, K.: Vegetation indices: advances made in biomass estimation and vegetation monitoring in the Last 30 years. Geocarto Int. 21, 21–28 (2006). doi: 10.1080/10106040608542399 CrossRefGoogle Scholar
  48. Singh, M., Evans, D., Tan, B.S., Nin, C.S.: Mapping and characterizing selected canopy tree species at the Angkor World Heritage Site in Cambodia using aerial data. PLoS One. 10, e0121558 (2015). doi: 10.1371/journal.pone.0121558 CrossRefPubMedPubMedCentralGoogle Scholar
  49. Thiam A.K.: Geographic Information Systems and Remote Sensing Methods for Assessing and Monitoring Land Degradation in the Sahel Region: The Case of Southern Mauritania (1998)Google Scholar
  50. Tucker, C.J.: Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 8, 127–150 (1979)CrossRefGoogle Scholar
  51. Vergara-Díaz, O., Zaman-Allah, M.A., Masuka, B., et al.: A novel remote sensing approach for prediction of maize yield under different conditions of nitrogen fertilization. Front. Plant Sci. (2016). doi: 10.3389/fpls.2016.00666
  52. Woebbecke, D.M., Meyer, G.E., Von Bargen, K., Mortensen, D.A.: Color Indices for Weed Identification Under Various Soil, Residue, and Lighting Conditions. Trans. ASAE. 38, 259–269 (1995). doi: 10.13031/2013.27838 CrossRefGoogle Scholar
  53. Zhang, F., Zaman, Q.U., Percival, D.C., Schumann, A.W.: Detecting bare spots in wild blueberry fields using digital color photography. Appl. Eng. Agric. 26, 723–728 (2010)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Biodiversity, Ecology and Evolution of Plants, Biocentre Klein Flottbek and Botanical GardenUniversity of HamburgHamburgGermany
  2. 2.Applied Landscape Ecology and Ecological Planning Institute of Landscape EcologyMünsterGermany
  3. 3.Faculty of Natural Resources and Spatial SciencesNamibia University of Science and TechnologyWindhoekNamibia

Personalised recommendations