Journal of Wetlands Research (한국습지학회지)

Korean Wetlands Society (한국습지학회)
권호 표지
DOI QR코드

DOI QR Code

A study for Improvement the Accuracy of Tree Species Classification within Various Sizes of Training Sample Areas by Using the High-resolution Images

고해상도 영상을 이용한 샘플영역의 크기별 수종분류 정확도 향상을 위한 연구

  • Hou, Jin Sung (Department of Civil & Environmental Engineering, Kongju National University) ;
  • Yang, Keum Chul (Department of Civil & Environmental Engineering, Kongju National University)
  • 허진성 (공주대학교 건설환경공학과) ;
  • 양금철 (공주대학교 건설환경공학과)
  • Received : 2014.03.21
  • Accepted : 2014.07.01
  • Published : 2014.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study was to investigate the objective impact in accuracy and reliability with tendency depend on training samples by using the high-resolution images. Supervised classification was performed based on multi-spectral images which made by each satellite and aerial images for considering all of bands' characteristics. The highest accuracy was 84.7% with satellite image(3*3) and 83% with aerial image(5*5) at the accuracy verification phase. Also, the overall accuracy with the consideration of Kappa coefficient were 0.84 for satellite images and 0.82 for aerial images. In all of the images, the smaller training sample was, the higher accuracy showed. Therefore, tree species classification accuracy was tended to rely on training sample size.

본 연구는 고해상도 위성영상과 항공영상을 이용하여 샘플영역의 크기 변화에 따른 수종분류 시 정확도와 신뢰도에 미치는 영향을 객관적으로 규명하고 그 경향성을 파악하는데 목적이 있다. 영상이 포함하고 있는 밴드들의 특성을 모두 고려하여 수종분류를 실시하기 위해 위성영상과 항공영상 각각에 대해 다중분광영상을 제작하였으며, 이를 기반으로 감독분류를 수행하였다. 그리고 정확도 검증단계에서 전체정확도를 산출하였으며, 그 결과 위성영상의 3*3에서 84.7%, 항공영상은 5*5에서 83%로 가장 높게 나타났으며 Kappa 계수는 각각 0.84, 0.82로 나타났다. 또한 두 영상의 샘플영역의 크기가 작아질수록 정확도가 높아지는 것으로 판단된다.

Keywords

References

  1. Brandtberg, T (2002). Individual tree-based species classification in high spatial resolution aerial images of forests using fuzzy sets. Fuzzy Sets and Systems 132, pp. 287-371
  2. Choi, HM, Yang, KC (2012). Classification of tree species using high-resolution QuickBird-2 satellite images in the valley of Ui-dong in Bukhansan National Park. J. of Ecology and Environment 35(2), pp. 91-98 https://doi.org/10.5141/JEFB.2012.013
  3. Chung, SY (2011). Study on the estimation of forest biomass using remote sensing data, Ph.D. Dissertation, Kookmin University, Seoul, Korea.
  4. Cohen, J (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement 20, pp. 37-46 https://doi.org/10.1177/001316446002000104
  5. ERDAS (1997). ERDAS Field Guide. 4th Edtion. Erdas Publisher, Atlanta, Georgia.
  6. Ha, SR, Kyoung, CK, Park, SY, Park, DH (2002). Estimation of classification Accuracy of JERS-1 satellite imagery according to the acquisition method and size of training reference data. J. of Korean Association of Geographic Information Studies 5(1), pp. 27-37. [Korean Literature]
  7. The Intergovernmental Panel on Climate Change (2007). Climate Change 2007. Syntheis Report. IPCC Fourth Assessment Report. Geneva, Switzerland, 104p
  8. Jang, AJ, Yu, KY, Kim, YI, Lee, BK ( 2006) . Estimation of individual tree and tree height using color aerial photograph and LiDAR data. J. of Korean Remote Sensing 22(6), pp. 543-551. [Korean Literature]
  9. Kang, NY, Park, JG, Cho, GS, Yeu, Y (2012). An analysis of land cover classification methods using IKONOS saellite Image. J. of the Korean Society for GeoSpatial Information System 20(3), pp. 65-71. [Korean Literature]
  10. Kwak, JH, Rhee, HJ, Lee, JW(1999). Feature selection of training set for supervised classification of satellite imagery. J. of the Korean Society of Remote Sensing 15(1), pp. 39-50 [Korean Literature]
  11. Katoh, M (2004). Classifying tree species in a northern mixed forest using high-resolution IKONOS data. The Japanese Forestry Society and Spring-Verlag Tokyo 9, pp. 7-14
  12. Kim, ES ( 2005). A study on improvement of classification accuracy using the texture information of high resolution panchromatic image, Master's Thesis, Inha University, Incheon, Korea, 53pp. [Korean Literature]
  13. Kwon, BK, Yamada, K, Niren, T, Jo, MH (2003). A study on the landcover classification using band rationing data of Landsat-TM. J. of Korean Association of Geographic Information Studies 6(2), pp. 80-91. [Korean Literature]
  14. Korea environment institute (2008). Application of Hyperspectral Image Data for Vegetation Analysis. KEI-2008-BA-06 , Korea environment institute. [Korean Literature]
  15. Lee, HJ, Lu, JH, Kim, SY (2011). Land cover object-oriented base classification using digital aerial photo image. J. of the Korean Society for GeoSpatial Information System 19(1), pp. 105-113. [Korean Literature]
  16. Nowongu (1999-2011). Nowon Statistical Yearbook. 51-3100000-000001-10, Nowongu, Seoul, Korea. [Korean Literature]
  17. Oh, CY, Park, SY, Kim, HS, Lee, YW, Choi, CU (2010). Comparison of landcover map accuracy using high resolution satellite imagery. J. of Korean Association of Geographic Information Studies 13(1), pp. 89-100. [Korean Literature]
  18. Park, JG, Go, SY, Cho, GS (2013). Vegetation classification using KOMPSAT-2 imagery and high-resolution airborne imagery in urban Area. J. of the Korean Society for GeoSpatial Information System 21(4), pp. 21-27. [Korean Literature]
  19. Whitesid, TG, Boggs, GS, Maier, SW (2011). Comparing object- bases and pixel-based classifications for mapping savannas. International J. of Applied Earth Observation and Geoinformation 13, pp. 884-893. https://doi.org/10.1016/j.jag.2011.06.008
  20. Zhang, Y (2004). Understanding image fusion. Photogrammetric Engineering and Remote Sensing 70(6), pp. 657-661.