Journal of the Korean Society of Environmental Restoration Technology (한국환경복원기술학회지)

The Korea Society of Environmental Restoration Technology (한국환경복원기술학회)
권호 표지
DOI QR코드

DOI QR Code

Evaluation of the Feasibility of Deep Learning for Vegetation Monitoring

딥러닝 기반의 식생 모니터링 가능성 평가

  • Received : 2023.10.27
  • Accepted : 2023.11.27
  • Published : 2023.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes a method for forest vegetation monitoring using high-resolution aerial imagery captured by unmanned aerial vehicles(UAV) and deep learning technology. The research site was selected in the forested area of Mountain Dogo, Asan City, Chungcheongnam-do, and the target species for monitoring included Pinus densiflora, Quercus mongolica, and Quercus acutissima. To classify vegetation species at the pixel level in UAV imagery based on characteristics such as leaf shape, size, and color, the study employed the semantic segmentation method using the prominent U-net deep learning model. The research results indicated that it was possible to visually distinguish Pinus densiflora Siebold & Zucc, Quercus mongolica Fisch. ex Ledeb, and Quercus acutissima Carruth in 135 aerial images captured by UAV. Out of these, 104 images were used as training data for the deep learning model, while 31 images were used for inference. The optimization of the deep learning model resulted in an overall average pixel accuracy of 92.60, with mIoU at 0.80 and FIoU at 0.82, demonstrating the successful construction of a reliable deep learning model. This study is significant as a pilot case for the application of UAV and deep learning to monitor and manage representative species among climate-vulnerable vegetation, including Pinus densiflora, Quercus mongolica, and Quercus acutissima. It is expected that in the future, UAV and deep learning models can be applied to a variety of vegetation species to better address forest management.

Keywords

Acknowledgement

본 논문은 환경부의 환경기술개발사업(과제번호: 2021003360001)의 지원을 받아 한국환경연구원이 수행한 "ICT 기반 생태계 모니터링 기술 및 동식물 탐지 AI 알고리즘 개발(2023-021R)" 사업의 연구결과로 작성되었으며, 일부 재인용이 되었음을 알립니다.

References

  1. Ali, S., Haixing, Z., Qi, M. Liang, S., Ning, J., Jia, Q., & Hou, F. (2021) Monitoring drought events and vegetation dynamics in relation to climate change over mainland China from 1983 to 2016. Environ Sci Pollut Res 28, 21910-21925. https://doi.org/10.1007/s11356-020-12146-4
  2. Bycroft, R., Leon, J. X., & Schoeman, D. (2019). Comparing random forests and convoluted neural networks for mapping ghost crab burrows using imagery from an unmanned aerial vehicle. Estuarine, Coastal and Shelf Science, 224, 84-93. https://doi.org/10.1016/j.ecss.2019.04.050
  3. Chen,Y., Guerschman,JP., Cheng,Z., & Guo.,L. (2019). Remote sensing for vegetation monitoring in carbon capture storage regions: A review,Applied Energy, Volume 240, 312-326. https://doi.org/10.1016/j.apenergy.2019.02.027
  4. Ferreira, M. P., Almeida, D., Papa, D., Minervino, J., Veras, H., Formighieri, A., Santos, C., Ferreira, M., Figueiredo, E., & Ferreira.E. (2020). Individual tree detection and species classification of Amazonian palms using UAV images and deep learning. Forest Ecology and Management, 475, 118397.
  5. Jang, K., (2021). A Study on the Deep Learning-based Tree Species Classification by using High-resolution Orthophoto Images. Journal of the Korean Association of Geographic Information Studies, 24(3), 1-9. (in Korean) https://doi.org/10.11108/KAGIS.2021.24.3.001
  6. Kattenborn, T., Eichel, J., & Fassnacht, F. E. (2019). Convolutional Neural Networks enable efficient, accurate and fine-grained segmentation of plant species and communities from high-resolution UAV imagery. Scientific reports, 9(1), 17656.
  7. Kattenborn, T., Leitloff, J., Schiefer, F., & Hinz, S. (2021). Review on Convolutional Neural Networks (CNN) in vegetation remote sensing. ISPRS journal of photogrammetry and remote sensing, 173, 24-49. https://doi.org/10.1016/j.isprsjprs.2020.12.010
  8. Kim, D., Yu, J., Yoon, J., Jeon, S., & Son, S. (2021). Comparison of accuracy of surface temperature images from unmanned aerial vehicle and satellite for precise thermal environment monitoring of urban parks using in situ data. Remote Sensing, 13(10), 1977.
  9. Kim, SH., Kwon, KW., & Kim HJ. (2022). A Study on Orthogonal Image Detection Precision Improvement Using Data of Dead Pine Trees Extracted by Period Based on U-Net model. ournal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography. 40(4), 251-260. (in Korean)
  10. Korea Forest Research Institute. (2010). Research on the Occurrence and Ecological Characteristics and Control of Oak Wilt Disease. (in Korean)
  11. Korean National Arboretum. (2010). 300 Target Plants Adaptable to Climate Change in the Korean Peninsula. Korea National Arboretum, Pocheon, Korea. 492. (in Korean)
  12. Korean National Arboretum. (2020). Climate Change Indicator Forest Plant Seasonal Observation Monitoring Manual, Pocheon, Korea. 262. (in Korean)
  13. Lee, SH., Lee, M. (2020). A Study on Deep Learning Optimization by Land Cover Classification Item Using Satellite Imagery. Korean Journal of Remote Sensing, 36(6-2), 1591-1604. (in Korean) https://doi.org/10.7780/KJRS.2020.36.6.2.9
  14. Ministry of Environment. (2023). Study on the Selection Criteria for National Environment Green Restoration Candidate List. (in Korean)
  15. National Institute of Ecology. (2019). Guidelines for the 5th National Ecosystem Survey. (in Korean)
  16. Park, Ju., Lee, M., & Choi, SY. (2021). Analysis of Trees Damaged by Pine Wilt Nematodes Using Unmanned Aerial Images. Journal of the Korean Cadastre Information Association, 23(2), 78-86. (in Korean) https://doi.org/10.46416/JKCIA.2021.08.23.2.78
  17. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, 234-241.
  18. Son, S. W., Kim, D. W., Sung, W. G., & Yu, J. J. (2020). Integrating UAV and TLS approaches for environmental management: A case study of a waste stockpile area. Remote Sensing, 12(10), 1615.
  19. Yu, J.J., Kim, D. W., Lee, E. J., & Son, S. W. (2022). Mid-and Short-Term Monitoring of Sea Cliff Erosion based on Structure-from-Motion (SfM) Photogrammetry: Application of Two Differing Camera Systems for 3D Point Cloud Construction. Journal of Coastal Research, 38(5), 1021- 1036.  https://doi.org/10.2112/JCOASTRES-D-21-00122.1