Performance of Artificial Neural Network for Tree Species Identification using Sentinel-2 Data

Year : 2024 | Volume :15 | Issue : 02 | Page : 13-22
By

Jaya Mini LWKN,

IAKS Illeperuma,

  1. Student Department of Remote Sensing and GIS, Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka
  2. Senior Lecturer Department of Remote Sensing and GIS, Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka

Abstract

Accurate land cover mapping, especially concerning vegetation, is crucial for effective land use policy planning and sustainable forest management. Hence, achieving accuracy in mapping requires a deep understanding of composition changes, vegetation conditions, and the spatial distribution of tree species. In the spatial context of tree species, it holds significant potential for applications including invasive species monitoring, delineating contaminated areas, and biodiversity conservation. However, traditional methods for tree species identification are often time-consuming and costly. Exploiting remote sensing technologies presents a promising solution to this challenge, offering the ability to classify vegetation over large areas. This study focuses on utilizing Sentinel-2 satellite data and artificial neural networks (ANNs) to classify four tree species in Elpitiya DS division, Sri Lanka: Tea (Camellia sinensis), Rubber (Hevea brasiliensis), Oil Palm (Elaeis guineensis), and Cinnamon (Cinnamomum verum). The ANN model achieved an overall accuracy of 78.33% in classifying these species, demonstrating the potential of Sentinel-2 images in such applications. However, accurate differentiation of rubber plantations proved challenging due to their small-scale nature and spectral similarities with other species. The research underscores the need for further model optimization to enhance accuracy and emphasizes the significance of satellite data for tree species identification. Despite promising results, continual refinement is essential to improve the model’s performance. Ultimately, this study highlights the potential of ANNs as a leading classification technique for land cover mapping and tree species identification.

Keywords: Artificial Neural Networks (ANNs), Elpitiya DS Division, Sentinel-2, Tree Species Identification

[This article belongs to Journal of Remote Sensing & GIS(jorsg)]

How to cite this article: Jaya Mini LWKN, IAKS Illeperuma. Performance of Artificial Neural Network for Tree Species Identification using Sentinel-2 Data. Journal of Remote Sensing & GIS. 2024; 15(02):13-22.
How to cite this URL: Jaya Mini LWKN, IAKS Illeperuma. Performance of Artificial Neural Network for Tree Species Identification using Sentinel-2 Data. Journal of Remote Sensing & GIS. 2024; 15(02):13-22. Available from: https://journals.stmjournals.com/jorsg/article=2024/view=167762

References

  1. Adelabu, S., et al., Exploiting machine learning algorithms for tree species classification in a semiarid woodland using RapidEye image. Journal of Applied Remote Sensing, 2013. 7(1): p. 073480-073480.
  2. Adiningrat, D.P., Mapping Dominant Tree Species from Remotely Sensed Image Using Machine Learning Algorithms. 2017, University of Twente.
  3. Agriculture, M.o.P.I.E., Statistical Information on Plantation Crops – 2018. 2020.
  4. org. Artificial Neural Network. 10:31, March 18, 2024; Available from: https://brilliant.org/wiki/artificial-neural- network/.
  5. Civco, D.L., Artificial neural networks for land- cover classification and mapping. International journal of geographical information science, 1993. 7(2): p. 173-186.
  6. Clark, M.L., Comparison of multi-seasonal Landsat 8, Sentinel-2 and hyperspectral images for mapping forest alliances in Northern California. ISPRS journal of photogrammetry and remote sensing, 2020. 159: p. 26-40.
  7. Dixon, B. and N. Candade, Multispectral landuse classification using neural networks and support vector machines: one or the other, or both? International Journal of Remote Sensing, 2008. 29(4): p. 1185-1206.
  8. Drusch, M., et al., Sentinel-2: ESA’s optical high-resolution mission for GMES operational services. Remote sensing of Environment, 2012. 120: p. 25-36.
  9. Follak, S., C. Schleicher, and M. Schwarz, Roads support the spread of invasive in Austria. Die Bodenkultur: Journal of Land Management, Food and Environment, 2018. 69(4): p. 257- 265.
  10. Gascon, F., et al., Copernicus Sentinel-2A calibration and products validation status. Remote Sensing, 2017. 9(6): p. 584.
  11. Hamad, R., An assessment of artificial neural networks, support vector machines and decision trees for land cover classification using sentinel-2A data. Sciences, 2020. 8(6): p. 459- 464.
  12. Hasan, M., et al., Comparative analysis of SVM, ANN and CNN for classifying vegetation species using hyperspectral thermal infrared data. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2019. 42: p. 1861-1868.
  13. Immitzer, M., C. Atzberger, and T. Koukal, Tree species classification with random forest using very high spatial resolution 8-band WorldView- 2 satellite data. Remote sensing, 2012. 4(9): p. 2661-2693.
  14. Lu, D. and Q. Weng, A survey of image classification methods and techniques for improving classification performance. International journal of Remote sensing, 2007. 28(5): p. 823-870.
  15. LUPPD, ed. Land Use and Land Cover of Sri Lanka: Land Use Policy Planning Department. 2019.
  16. Mas, J.F. and J.J. Flores, The application of artificial neural networks to the analysis of remotely sensed data. International Journal of Remote Sensing, 2008. 29(3): p. 617-663.
  17. Móricz, N., et al., Modelling the potential distribution of three climate zonal tree species for present and future climate in Hungary= Három klímazonális fafaj hazai potenciális elterjedésének modellezése jelenlegi és jövőbeni klímában. Acta Silvatica et Lignaria Hungarica, 2013. 9(1): p. 85-96.
  18. Nitze, I., U. Schulthess, and H. Asche, Comparison of machine learning algorithms random forest, artificial neural network and support vector machine to maximum likelihood for supervised crop type classification. Proceedings of the 4th GEOBIA, Rio de Janeiro, Brazil, 2012. 79: p. 3540.
  19. Omer, G., et al., Performance of support vector machines and artificial neural network for mapping endangered tree species using WorldView-2 data in Dukuduku forest, South Africa. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015. 8(10): p. 4825-4840.
  20. Phiri, D., et al., Sentinel-2 data for land cover/use mapping: A review. Remote Sensing, 2020. 12(14): p. 2291.
  21. Raczko, E. and B. Zagajewski, Tree species classification of the UNESCO man and the biosphere Karkonoski National Park (Poland) using artificial neural networks and APEX hyperspectral images. Remote Sensing, 2018. 10(7): p. 1111.
  22. Rujoiu-Mare, M.-R., et al., Land cover classification in Romanian Carpathians and Subcarpathians using multi-date Sentinel-2 remote sensing imagery. European Journal of Remote Sensing, 2017. 50(1): p. 496-508.
  23. Soleimannejad, L., et al., Evaluating the potential of sentinel-2, landsat-8, and irs satellite images in tree species classification of hyrcanian forest of iran using random forest. Journal of sustainable forestry, 2019. 38(7): p. 615-628.
  24. Verikas, A., A. Gelzinis, and M. Bacauskiene, Mining data with random forests: A survey and results of new tests. Pattern recognition, 2011. 44(2): p. 330-349.
  25. Xi, Y., et al., Exploitation of time series sentinel-2 data and different machine learning algorithms for detailed tree species classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021. 14: p. 7589-7603.
Regular Issue Subscription Original Research
Volume 15
Issue 02
Received June 4, 2024
Accepted July 7, 2024
Published July 10, 2024
Views: 26

Check Our other Platform for Workshops in the field of AI, Biotechnology & Nanotechnology.
Learn more
Check Out Platform for Webinars in the field of AI, Biotech. & Nanotech.
Learn more