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

A Comparative Study of Image Classification Method to Detect Water Body Based on UAS

UAS 기반의 수체탐지를 위한 영상분류기법 비교연구

  • LEE, Geun-Sang (Dept. of Cadastre and Civil Engineering, Vision University of Jeonju) ;
  • KIM, Seok-Gu (Geospatial Information Co. Ltd.) ;
  • CHOI, Yun-Woong (Dept. of Civil Engineering, Chosun College of Science & Technology)
  • 이근상 (전주비전대학교 지적토목학과) ;
  • 김석구 ((주)공간정보) ;
  • 최연웅 (조선이공대학교 토목건설과)
  • Received : 2015.07.10
  • Accepted : 2015.09.04
  • Published : 2015.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, there has been a growing interest in UAS(Unmanned Aerial System), and it is required to develop techniques to effectively detect water body from the recorded images in order to implement flood monitoring using UAS. This study used a UAS with RGB and NIR+RG bands to achieve images, and applied supervised classification method to evaluate the accuracy of water body detection. Firstly, the result for accuracy in water body image classification by RGB images showed high Kappa coefficients of 0.791 and 0.783 for the artificial neural network and minimum distance method respectively, and the maximum likelihood method showed the lowest, 0.561. Moreover, in the evaluation of accuracy in water body image classification by NIR+RG images, the magalanobis and minimum distance method showed high values of 0.869 and 0.830 respectively, and in the artificial neural network method, it was very low as 0.779. Especially, RGB band revealed errors to classify trees or grasslands of Songsan amusement park as water body, but NIR+RG presented noticeable improvement in this matter. Therefore, it was concluded that images with NIR+RG band, compared those with RGB band, are more effective for detection of water body when the mahalanobis and minimum distance method were applied.

최근 무인비행시스템(UAS)에 대한 관심이 높아지고 있으며, 홍수시 UAS를 활용하여 침수모니터링을 수행하기 위해서는 촬영된 영상으로부터 수체를 효과적으로 탐지할 수 있는 기법 개발이 필요하다. 본 연구에서는 RGB와 NIR+RG 밴드를 탑재한 UAS를 활용하여 영상을 획득하였으며, 감독분류기법을 적용하여 수체탐지 정확도를 평가하였다. 먼저 RGB 영상에 의한 수체의 영상분류 정확도 평가에서는 인공신경망과 최소거리법의 Kappa 계수가 각각 0.791과 0.783로 높게 나타났으며, 최대우도법이 0.561로 가장 낮게 나타났다. 또한 NIR+RG 영상에 의한 수체의 영상분류 정확도 평가에서는 Mahalanobis와 최소거리법이 각각 0.869와 0.830으로 높게 나타났으며, 인공신경망법이 0.779로 매우 낮게 나타났다. 특히 RGB 밴드에서는 송산유원지의 수목이나 초지가 수체로 오분류되는 문제가 발생하였으나, NIR+RG 밴드에서는 이러한 문제가 많이 개선되었다. 따라서, RGB 밴드에 비해 NIR+RG 밴드를 탑재한 영상이 Mahalanobis와 최소거리법을 적용시 수체를 탐지하는데 효과적인 것으로 나타났다.

Keywords

References

  1. Aly, S. 2014. Learning invariant local image descriptor using convolutional mahalanobis self-organising map. Neurocomputing 142:239-247. https://doi.org/10.1016/j.neucom.2014.03.060
  2. Bendig, J., K. Yu, H. Aasen, A. Bolten, S. Benertz, J. Broscheit, M.L. Gnyp and G. Bareth. 2015. Combining UAV-based plant height from crop surface models, visible, and near infrared vegetation indices for biomass monitoring in barley. International Journal of Applied Earth Observation and Geoinformation 39:79-87. https://doi.org/10.1016/j.jag.2015.02.012
  3. Choi, K.A. and I.P. Lee. 2009. Image georeferencing using AT without GCPs for a UAV-based low-cost multisensor system. Korean Journal of Geomatics 27(2):249-260(최경아, 이임평. 2009.UAV기반 저가 멀티센서시스템을 위한 무기준점 AT를 이용한 영상의 Georeferencing.한국측량학회지 27(2):249-260).
  4. Choi, K.A., J.H. Lee and I.P. Lee. 2011. Development of a close-range realtime aerial monitoring system based on a low altitude unmanned air vehicle. Journal of Korea Spatial Information Society 19(4):21-31(최경아, 이지훈, 이임평. 2011. 저고도 무인 항공기 기반의 근접 실시간 공중 모니터링 시스템 구축. 한국공간정보학회지 19(4):21-31).
  5. Colomina, I., M. Blazquez, P. Molina, M.E. Pares and M. Wis. 2008. Towards a new paradigm for high-resolution lowcost photogrammetry and remote sensing. ISPRS XXI Congress. Beijing, China, Jul. 3-11, 2008. pp.1201-1206.
  6. Fauvel, M., J. Chanussort, J.A. Benediktsson and A. Villa. 2013. Parsimonious mahalanobis kernel for the classification of high dimensional data, Pattern Recognition 46:845-854. https://doi.org/10.1016/j.patcog.2012.09.009
  7. George, H.M. and I.Z. Gitas. 2013. Mapping post-fire forest regeneration and vegetation recovery using a combination of very high spatial resolution and hyperspectral satellite imagery. International Journal of Applied Earth Observation and Geoinformation 20:60-66. https://doi.org/10.1016/j.jag.2011.09.001
  8. Han, J.H., Kwon, J.H., Lee, I. and Choi, K. 2010. Determinate real-time position and attitude using GPS/INS/AT for real-time aerial monitoring system. Korean Journal of Geomatics 28(5):531-537(한중희, 권재현, 이임평, 최경아. 2010. 실시간 공중 자료획득 시스템을 위한 GPS/INS/AT를 이용한 실시간 위치/자세 결정.한국측량학회지 28(5):531-537).
  9. Hong, S.Y., Rim, S.K., Lee, S.H., Lee, J.C. and Kim, Y.H. 2008. Spatial analysis of agro-environment of North Korea using remote sensing. Korean Journal of Environmental Agriculture 27(2):120-132(홍석영, 임상규, 이승호, 이정철, 김이현. 2008. 위성영상을 이용한 북한의 농업환경 분석. 한국환경농학회지 27(2):120-132). https://doi.org/10.5338/KJEA.2008.27.2.120
  10. Jang, D.H. 2005. Change detection of deposits environment in stream channel before and after flooding using high resolution satellite imagery : a case in the Sacheoncheon. Journal of the Korean Geomorphological Association 12(3):49-58(장동호. 2005. 고해상도 위성영상을 이용한 홍수 전후의 하도 내 퇴적환경 변화탐지 : 강릉 사천천 사례연구. 한국지형학회지 12(3):49-58).
  11. Juan, I., C.J.F. Ortega, D. Hernandez and M.A. Moreno. 2013. Estimation of leaf area index in onion using an unmanned aerial vehicle. Biosystems Engineering 115:31-42. https://doi.org/10.1016/j.biosystemseng.2013.02.002
  12. Jung, S.H., H.M. Lim and J.K. Lee. 2010. Acquisition of 3D spatial information using UAV photogrammetric method. Korean Journal of Geomatics 28(1):161-168(정성혁, 임형민, 이재기. 2010. 무인항공 사진측량을 이용한 3D 공간정보 취득.한국측량학회지 28(1):161-168).
  13. Kang, M.S., S.W. Park and K.S. Yoon. 2006. Land cover classification of image data using artificial neural networks. Rural Planning 12(1):75-83(강문성, 박승우, 윤광식. 2006. 인공신경망 모형을 이용한 영상자료의 토지피복분류. 농촌계획 12(1):75-83).
  14. Kim, S.H., S.J. Kang and K.S. Lee. 2010. Multi-temporal Landsat ETM+ mosaic method for generating land cover map over the Korean peninsula. Korean Journal of Remote Sensing 26(2):87-98(김선화, 강성진, 이규성. 2010. 한반도 토지피복도 제작을 위한 다시기 LandsatETM+ 영상의 정합 방법. 대한원격탐사학회지 26(2):87-98). https://doi.org/10.7780/kjrs.2010.26.2.87
  15. Kim, S.M. 2014. Study of the UAV for application plans and landscape analysis. Journal of the Korean Institute of Traditional Landscape Architecture 32(3):213-220(김승민. 2014. UAV를이용한 경관분석 및 활용방안에 관한 기초연구. 한국전통조경학회지 32(3):213-220). https://doi.org/10.14700/KITLA.2014.32.3.213
  16. Kim, S.S. 2007. Flood hazard analysis by imagery acquisition system based on an unmanned helicopter, Ph.D. Thesis, Gyeongsang National University, Jinju, Korea. (김성삼. 2007. 무인헬기기반 영상정보취득체계에 의한 홍수재해 분석. 경상대학교 대학원 박사학위논문).
  17. Kim, Y.S., Seo, E.S. and Cho, M.H. 1998. The Concept of Remote Sensing. Donghwa Technology Corporation (김영섭, 서애숙, 조명희. 1998. 원격탐사개론. 동화기술. 241-245쪽).
  18. Kuchenhoff, H., T. Augustin and A. Kunz. 2012. Partially identified prevalence estimation under misclassification using the kappa coefficient. International Journal of Approximate Reasoning. 53(8) :1168-1182. https://doi.org/10.1016/j.ijar.2012.06.013
  19. Lee, G.S. and I.Y. Jeong. 2008. Development of 3D GIS system for the visualization of flood inundation area. Journal of Korean Society of Civil Engineering 28(5D):749-757 (이근상,정일영. 2008. 홍수범람지역 가시화를 위한3차원 GIS 시스템 개발. 대한토목학회논문집 28(5D):749-757).
  20. Lee, G.S. and Y.W. Choi. 2014. Land cover classification of Nakdong basin using object-based analysis methods. Journal of the Korean Cadastre Information Association 16(3):3-18 (이근상, 최연웅. 2014. 객체기반 영상분석 기법에 의한 낙동강 유역 토지피복 분류. 한국지적정보학회지 16(3):3-18).
  21. Lee, H.R. and H.M. Kim. 2005. A correlation analysis between land surface temperature and NDVI in Kunsan city using Landsat 7 TM/ETM+satellite images. Journal of the Korean Association of Geographic Information Studies 8(2):31-43 (이홍로, 김형무.2005. Landsat 7 TM ETM 위성영상을 이용한 군산지역 지표 온도와 NDVI에 대한상관분석. 한국지리정보학회지 8(2):31-43).
  22. Lin, C.S. and C.S. Lee. 1995. Neural Fuzzy Systems. Prentice Hill, pp.236-244.
  23. Mielke, P.W. and K.J. Berry. 2009. A note on Cohen's weighted kappa coefficient of agreement with linear weights. Statistical Methodology 6(5):439-446. https://doi.org/10.1016/j.stamet.2009.03.002
  24. Mumby, P.J., E.P. Green, A.J. Edwards and C.D. Clark. 1997. Coral reef habitat-mapping : how much detail can remote sensing provide?. Marine Biology 1997:193-202.
  25. Nagai, M., T. Chen and A. Ahmed and R. Shibasaki. 2008. UAV borne mapping by multi sensor integration. ISPRS XXI Congress. Beijing, China, Jul. 3-11, 2008, pp.1215-1221.
  26. National Disaster Management Institute. 2007. Development of damage estimation module using geospatial imagery information (소방방재청. 2007. 공간영상 정보를 이용한 피해추정액 산정 및 모듈개발).
  27. Nebiker, S., A. Annen, M. Scherrer and D. Oesch. 2008. A light-weight multispectral sensor for micro UAV - opportunities for very high resolution airborne remote sensing. ISPRS XXI Congress. Beijing, China, Jul. 3-11, 2008. pp.1193-1199.
  28. Olmanson, L.G., L.P. Brezonik and M.E. Bauer. 2013. Airborne hyperspectral remote sensing to assess spatial distribution of water quality characteristics in large rivers: the Mississippi river and its tributaries in Minnesota. Remote Sensing of Environment 130:254-265. https://doi.org/10.1016/j.rse.2012.11.023
  29. Park, S.M., J.H. Im and H.S. Sakong. 2001. Land cover classification of a wide area through multi-scene Landsat processing. Korean Journal of Remote Sensing 17(3):189-197 (박성미, 임정호,사공호상. 2001. 다량의 Landsat 위성영상처리를 통한 광역 토지피복분류. 대한원격탐사학회지 17(3):189-197).
  30. Pasqualini, V., C. Pergent-Martine, C. Fernandez and G. Pergent. 1997. The use of airborne remote sensing for benthic cartography : advantages and reliability. International Journal of Remote Sensing 18(5):1167-1177. https://doi.org/10.1080/014311697218638
  31. Sakong, H.S. and J.H. Rim. 2003. An empirical study on the land cover classification method using IKONOS image. Journal of the Korean Association of Geographic Information Studies 6(3):107-116 (사공호상, 임정호. 2003.IKONOS 영상의 토지피복분류 방법에 대한실증 연구. 한국지리정보학회지 6(3):107-116).
  32. Schalkoff, R.J. 1992. Pattern Recognition Statistical, Structural and Neural Approaches. John Wiley & Sons, Inc.
  33. Yoo, H.H., J.W. Park, J.H. Shim and S.S. Kim. 2006. Image map generation using low-altitude photogrammetric UAV. Journal of The Korean Society for Geo-Spatial Information System 14(1): 37-47 (유환희, 박장환, 심재현, 김성삼.2006. 저고도촬영시스템을 이용한 영상지도제작, 한국지형공간정보학회지 14(1):37-47).
  34. Yoon, S.T., O.S. Kim and S.K. Rim. 2001. Evaluation of land cover classification of Pyeong-Taeg area by Landsat thematic mapper data. Korean Journal of Agricultural and Forest Meteorology 3(3):163-170 (윤성탁, 김선오, 임상규.2001. Landsat TM 영상자료를 이용한 평택지역의 토지피복 현황 및 분류정확도 평가. 한국농림기상학회지 3(3):163-170).

Cited by

  1. Estimation of Reservoir Area and Capacity Curve Equation using UAV Photogrammetry vol.24, pp.3, 2016, https://doi.org/10.7319/kogsis.2016.24.3.093
  2. Orthophoto and DEM Generation in Small Slope Areas Using Low Specification UAV vol.34, pp.3, 2016, https://doi.org/10.7848/ksgpc.2016.34.3.283
  3. Test of Fault Detection to Solar-Light Module Using UAV Based Thermal Infrared Camera vol.19, pp.4, 2016, https://doi.org/10.11108/kagis.2016.19.4.106
  4. A Comparative Study of Image Classification Method to Classify Onion and Garlic Using Unmanned Aerial Vehicle (UAV) Imagery vol.49, pp.6, 2016, https://doi.org/10.7745/KJSSF.2016.49.6.743
  5. UAV 기반 열적외선 센서를 이용한 태양광 셀의 발열 검출 vol.25, pp.1, 2017, https://doi.org/10.7319/kogsis.2017.25.1.071
  6. Selection of Optimal Vegetation Indices and Regression Model for Estimation of Rice Growth Using UAV Aerial Images vol.50, pp.5, 2015, https://doi.org/10.7745/kjssf.2017.50.5.409
  7. Development of a Semi-automatic Search Program for Crown Delineation Based on Watershed and Valley Following Algorithms vol.34, pp.2, 2015, https://doi.org/10.7747/jfes.2018.34.2.142
  8. 식생 활력도를 고려한 드론 기반의 식생지수 분석 vol.23, pp.2, 2015, https://doi.org/10.11108/kagis.2020.23.2.021
  9. A Study on Obtaining Tree Data from Green Spaces in Parks Using Unmanned Aerial Vehicle Images: Focusing on Mureung Park in Chuncheon vol.24, pp.4, 2021, https://doi.org/10.11628/ksppe.2021.24.4.441