Journal of the Korean Association of Geographic Information Studies (한국지리정보학회지)

The Korean Association of Geographic Information Studies (한국지리정보학회)
권호 표지
DOI QR코드

DOI QR Code

Classification of Urban Green Space Using Airborne LiDAR and RGB Ortho Imagery Based on Deep Learning

항공 LiDAR 및 RGB 정사 영상을 이용한 딥러닝 기반의 도시녹지 분류

  • SON, Bokyung (Dept. of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology) ;
  • LEE, Yeonsu (Dept. of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology) ;
  • IM, Jungho (Dept. of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology)
  • 손보경 (울산과학기술원 도시환경공학과) ;
  • 이연수 (울산과학기술원 도시환경공학과) ;
  • 임정호 (울산과학기술원 도시환경공학과)
  • Received : 2021.09.10
  • Accepted : 2021.09.24
  • Published : 2021.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Urban green space is an important component for enhancing urban ecosystem health. Thus, identifying the spatial structure of urban green space is required to manage a healthy urban ecosystem. The Ministry of Environment has provided the level 3 land cover map(the highest (1m) spatial resolution map) with a total of 41 classes since 2010. However, specific urban green information such as street trees was identified just as grassland or even not classified them as a vegetated area in the map. Therefore, this study classified detailed urban green information(i.e., tree, shrub, and grass), not included in the existing level 3 land cover map, using two types of high-resolution(<1m) remote sensing data(i.e., airborne LiDAR and RGB ortho imagery) in Suwon, South Korea. U-Net, one of image segmentation deep learning approaches, was adopted to classify detailed urban green space. A total of three classification models(i.e., LRGB10, LRGB5, and RGB5) were proposed depending on the target number of classes and the types of input data. The average overall accuracies for test sites were 83.40% (LRGB10), 89.44%(LRGB5), and 74.76%(RGB5). Among three models, LRGB5, which uses both airborne LiDAR and RGB ortho imagery with 5 target classes(i.e., tree, shrub, grass, building, and the others), resulted in the best performance. The area ratio of total urban green space(based on trees, shrub, and grass information) for the entire Suwon was 45.61%(LRGB10), 43.47%(LRGB5), and 44.22%(RGB5). All models were able to provide additional 13.40% of urban tree information on average when compared to the existing level 3 land cover map. Moreover, these urban green classification results are expected to be utilized in various urban green studies or decision making processes, as it provides detailed information on urban green space.

도시녹지는 도시 생태계 건강성 증진을 위한 중요한 요소이며, 건강한 도시 생태계 유지 및 관리를 위해서는 도시녹지의 공간적인 현황 파악이 필요하다. 환경부에서는 2010년 이후부터 총 41개의 분류 항목을 갖는 1m 급 해상도의 세분류 토지피복지도를 제공해오고 있으나, 가로수와 같은 도시 내 고해상도 상세 녹지 정보는 기타 초지로 분류되거나 누락되어 오고 있다. 따라서, 본 연구에서는 수원시 지역을 대상으로 1m 이하 급의 고해상도 원격탐사 자료(항공 LiDAR 및 RGB 정사영상)를 이용하여, 기존 세분류 토지피복지도에서는 나타나지 않는 고해상도의 상세 도시 녹지(수목, 관목 및 초지) 정보를 분류하고자 하였다. 분류 기법으로는 딥러닝 기반의 이미지 분할방법인 U-Net 구조의 모델을 활용하였으며, 분류 항목의 수 및 사용하는 자료의 종류에 따라 총 3가지의 모델(LRGB10, LRGB5, 및 RGB5)을 제안하고 성능을 평가하였다. 검증 지역에 대한 세 모델의 평균 전체 정확도는 각 83.40%(LRGB10), 89.44%(LRGB5), 74.76%(RGB5)이며, 항공 LiDAR와 RGB 정사영상을 함께 사용하여 총 5개의 항목(수목, 관목, 초지, 건물, 및 그 외)을 분류하는 LRGB5 모델의 성능이 가장 높게 나타났다. 수원시의 수목, 관목 및 초지 기준의 전체 녹지 현황은 각 45.61%(LRGB10), 43.47%(LRGB5), 및 44.22%(RGB5)로 나타났으며, 세 모델 모두 기존 세분류 토지피복지도와 비교하여 평균 13.40%의 도시 수목 정보를 더 제공할 수 있는 것으로 나타났다. 더불어 이러한 도시녹지 분류 결과는 향후 중분류 토지피복지도와 같은 기존 GIS 정보와의 융합을 통해 가로수 녹지 비율 현황 등 추가적인 상세 녹지 현황 정보를 제공할 수 있어, 다양한 도시녹지 연구 및 정책의 기초 자료로 활용될 수 있을 것으로 기대된다.

Keywords

Acknowledgement

본 연구는 환경부의 재원으로 한국환경산업기술원의 도시생태 건강성 증진 기술개발사업의 지원을 받아 수행되었음(2020002770001).

References

  1. Armson, D., P. Stringer and A.R. Ennos. 2012. The effect of tree shade and grass on surface and globe temperatures in an urban area. Urban Forestry & Urban Greening 11(3):245-255. https://doi.org/10.1016/j.ufug.2012.05.002
  2. Cao, K. and X. Zhang. 2020. An improved resunet model for tree species classification using airborne high-resolution images. Remote Sensing 12(7):1128. https://doi.org/10.3390/rs12071128
  3. Chang, A.J., Y.I. Kim, B.K. Lee and K.Y. Yu. 2006. Estimation of individual tree and tree height using color aerial photograph and LiDAR data. Korean Journal of Remote Sensing 22(6):543-551. https://doi.org/10.7780/KJRS.2006.22.6.543
  4. Degerickx, J., M. Hermy and B. Somers. 2020. Mapping functional urban green types using high resolution remote sensing data. Sustainability 12(5):2144. https://doi.org/10.3390/su12052144
  5. Govender, M., K. Chetty and H. Bulcock. 2007. A review of hyperspectral remote sensing and its application in vegetation and water resource studies. Water Sa 33(2):145-151.
  6. Hartling, S., V. Sagan, P. Sidike, M. Maimaitijiang and J. Carron. 2019. Urban tree species classification using a WorldView-2/3 and LiDAR data fusion approach and deep learning. Sensors 19(6):1284. https://doi.org/10.3390/s19061284
  7. Hoeser, T. and C. Kuenzer. 2020. Object detection and image segmentation with deep learning on earth observation data: A review-part i: Evolution and recent trends. Remote Sensing. 12(10):1667. https://doi.org/10.3390/rs12101667
  8. Im, J., J.R. Jensen and M.E. Hodgson. 2008. Object-based land cover classification using high-posting-density LiDAR data. GIScience & Remote Sensing 45(2):209-228. https://doi.org/10.2747/1548-1603.45.2.209
  9. Jeanjean, A.P., P.S. Monks and R.J. Leigh. 2016. Modelling the effectiveness of urban trees and grass on PM2. 5 reduction via dispersion and deposition at a city scale. Atmospheric Environment 147:1-10. https://doi.org/10.1016/j.atmosenv.2016.09.033
  10. Jo, H.K., H.M. Park and J.Y. Kim. 2020. Carbon reduction and enhancement for greenspace in institutional lands. Journal of the Korean Institute of Landscape Architecture 48(4):1-7. https://doi.org/10.9715/KILA.2020.48.4.001
  11. Kim, M., J. Lee, D. Han, M. Shin, J. Im, J. Lee, L.J. Quackenbush and Z. Gu. 2018a. Convolutional neural network-based land cover classification using 2-D spectral reflectance curve graphs with multitemporal satellite imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 11(12):4604-4617. https://doi.org/10.1109/jstars.2018.2880783
  12. Kim, S.H., E.J. Park and I.K. Kim. 2018b. Estimation of carbon sequestration in urban green spaces using environmental spatial information - a case study of Ansan city -. Journal of the Korean Society of Environmental Restoration Technology 21(3):13-26.
  13. Lee, B.G. and Y.W. Lee. 2018. A study on utilization of spatial information for urban parks and green areas management. Korean Association Of Cadastre Information 20(2):151-162. https://doi.org/10.46416/JKCIA.2018.08.20.2.151
  14. Lee, S. and J. Kim. 2019. Land cover classification using sematic image segmentation with deep learning. Korean Journal of Remote Sensing 35(2):279-288. https://doi.org/10.7780/KJRS.2019.35.2.7
  15. Liu, L., N.C. Coops, N.W. Aven and Y. Pang. 2017. Mapping urban tree species using integrated airborne hyperspectral and LiDAR remote sensing data. Remote Sensing of Environment 200:170-182. https://doi.org/10.1016/j.rse.2017.08.010
  16. Lo, A.Y., J.A. Byrne and C.Y. Jim. 2017. How climate change perception is reshaping attitudes towards the functional benefits of urban trees and green space: lessons from Hong Kong. Urban Forestry & Urban Greening 23:74-83. https://doi.org/10.1016/j.ufug.2017.03.007
  17. Oh, C.Y., S.Y. Park, H.S. Kim, Y.W. Lee and C.U. Choi. 2010. Comparison of landcover map accuracy using high resolution satellite imagery. Journal of the Korean Association of Geographic Information Studies 13(1):89-100. https://doi.org/10.11108/KAGIS.2010.13.1.089
  18. Park, J.H. 1992. The status and legal system of the urban green space. Landscaping Tree 6(1_2):22-31.
  19. Park, S.C., B.H. Han, M.J. Park, H.D. Yun and M.J. Kim. 2016. A study on the possibility of utilizing both biotope maps and land cover maps on the calculation of the ecological network indicator of city biodiversity index. Journal of the Korean Institute of Landscape Architecture 44(6):73-83. https://doi.org/10.9715/KILA.2016.44.6.073
  20. Ronneberger, O., P. Fischer, and T. Brox. 2015. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, pp.234-241.
  21. Song, B.G. and K.H. Park. 2015. An analysis of rational green area ratio by land use types for mitigating heat-island effects. Journal of the Korean Association of Geographic Information Studies 18(2):59-74. https://doi.org/10.11108/kagis.2015.18.2.059
  22. Sothe, C., C.M. De Almeida, M.B. Schimalski, L.E.C. La Rosa, J.D.B. Castro, R.Q. Feitosa, M. Dalponte, C.L. Lima, V. Liesenberg, G.T. Miyoshi and A.M.G. Tommaselli. 2020. Comparative performance of convolutional neural network, weighted and conventional support vector machine and random forest for classifying tree species using hyperspectral and photogrammetric data. GIScience & Remote Sensing 57(3):369-394. https://doi.org/10.1080/15481603.2020.1712102
  23. Su, L. and J. Gibeaut. 2017. Using UAS hyperspatial RGB imagery for identifying beach zones along the South Texas Coast. Remote Sensing 9(2):159. https://doi.org/10.3390/rs9020159
  24. Sun, Y., J. Huang, Z. Ao, D. Lao and Q. Xin. 2019. Deep learning approaches for the mapping of tree species diversity in a tropical wetland using airborne LiDAR and high-spatial-resolution remote sensing images. Forests 10(11):1047. https://doi.org/10.3390/f10111047
  25. Sung, H.C. and S.H. Min. 2003. An empirical study on the function and effect of urban openspace - focusing on urban roadside trees -. Journal of the Korean Institute of Landscape Architecture 31(2):48-57.
  26. Suwon city. 2016. Suwon-si Environment conservation plan (2016~2025). https://www.suwon.go.kr/sw-www/deptHome/dep_env/env_01/env_01_03_02.jsp. (Accessed August 31, 2021).
  27. Thaiutsa, B., L. Puangchit, R. Kjelgren and W. Arunpraparut. 2008. Urban green space, street tree and heritage large tree assessment in Bangkok, Thailand. Urban forestry & urban greening 7(3):219-229. https://doi.org/10.1016/j.ufug.2008.03.002
  28. Van Thinh, T., P.C. Duong, K.N. Nasahara and T. Tadono. 2019. How does land use/land cover map's accuracy depend on number of classification classes?. SOLA 15:28-31. https://doi.org/10.2151/sola.2019-006
  29. Yoo, C., Y. Lee, D. Cho, J. Im and D. Han. 2020a. Improving local climate zone classification using incomplete building data and Sentinel 2 images based on convolutional neural networks. Remote Sensing 12(21):3552. https://doi.org/10.3390/rs12213552
  30. Yoo, S.H., J.S. Lee, J.S. Bae and H.G. Sohn. 2020b. Automatic generation of land cover map using residual U-Net. Journal of the Korean Society of Civil Engineers 40(5):535-546. https://doi.org/10.12652/KSCE.2020.40.5.0535
  31. Zhang, C., K. Xia, H. Feng, Y. Yang and X. Du. 2020. Tree species classification using deep learning and RGB optical images obtained by an unmanned aerial vehicle. Journal of Forestry Research pp.1-10.
  32. Zhong, Y., Q. Cao, J. Zhao, A. Ma, B. Zhao and L. Zhang. 2017. Optimal decision fusion for urban land-use/land-cover classification based on adaptive differential evolution using hyperspectral and LiDAR data. Remote Sensing 9(8):868. https://doi.org/10.3390/rs9080868