DOI QR코드

DOI QR Code

A Comparison of Pixel- and Segment-based Classification for Tree Species Classification using QuickBird Imagery

QuickBird 위성영상을 이용한 수종분류에서 픽셀과 분할기반 분류방법의 정확도 비교

  • Chung, Sang Young (Department of Forest Environment System, Kookmin University) ;
  • Yim, Jong Su (Division of Forest Resource Information, Korea Forest Research Institute) ;
  • Shin, Man Yong (Department of Forest Environment System, Kookmin University)
  • 정상영 (국민대학교 산림환경시스템학과) ;
  • 임종수 (국립산림과학원 산림자원정보과) ;
  • 신만용 (국민대학교 산림환경시스템학과)
  • Received : 2011.04.06
  • Accepted : 2011.10.04
  • Published : 2011.12.31

Abstract

This study was conducted to compare classification accuracy by tree species using QuickBird imagery for pixel- and segment-based classifications that have been mostly applied to classify land covers. A total of 398 points was used as training and reference data. Based on this points, the points were classified into fourteen land cover classes: four coniferous and seven deciduous tree species in forest classes, and three non-forested classes. In pixel-based classification, three images obtained by using raw spectral values, three tasseled indices, and three components from principal component analysis were produced. For the both classification processes, the maximum likelihood method was applied. In the pixel-based classification, it was resulted that the classification accuracy with raw spectral values was better than those by the other band combinations. As resulted that, the segment-based classification with a scale factor of 50% provided the most accurate classification (overall accuracy:76% and ${\hat{k}}$ value:0.74) compared to the other scale factors and pixel-based classification.

본 연구는 고해상도 위성영상인 QuickBird 영상을 이용한 픽셀 및 분할기반의 분류방법의 정확도를 비교하여 적합한 수종 분류방법을 선정하기 위해 수행하였다. 이를 위해 연구대상지인 충청북도 옥천군과 영동군의 산림을 대상으로 현지조사를 실시하여 총 398개 토지피복정보를 수집하였다. 총 14개의 토지 피복 등급(4개의 침엽수종과 7개의 활엽수종, 그리고 3개의 비산림지)으로 구분된 현지조사 자료를 훈련자료로 이용하였다. 픽셀기반 분류에 있어서 위성영상이 가지고 있는 원 화소값, tasseled cap 분석에 의한 3개의 지수, 그리고 주성분 분석을 통한 3개의 성분값을 이용한 3가지의 밴드조합 영상을 생성하여 분류정확도를 비교한 결과, 위성영상의 원 화소값을 이용한 분류 정확도가 가장 높은 것으로 평가되었다. 분할기반 분류에서는 3개의 축척계수에 따른 정확도를 비교한 결과, 축척계수 50%을 적용하였을 때 전체 정확도는 76%, 그리고 ${\hat{k}}$ 값은 0.74로 다른 축척계수에 의한 정확도보다 높은 것으로 나타났다. 결과적으로 QuickBird 영상의 원 화소값과 50%의 축척계수를 이용한 분할기반의 수종분류 결과가 정확도가 가장 높은 것으로 평가되었다.

Keywords

References

  1. Battz, M. and Schape, A. 2000. Multiresolution Segmentation: An Optimization Approach for High Quality Multi-scale Image Segmentation. Proceedings of the 12th Symposium for Applied Geographic Information Processing. Salzburg, Austria, 12-23.
  2. Benz, U. 2001. Definiens Imaging GmbH:Object-Oriented Classification and Feature Detection. IEEE Geoscience and Remote Sensing Society Newsletter 16-20.
  3. Blaschke, T. and Strobl, J. 2001. What's wrong with Pixels? Some Recent Developments Interfacing Remote Sensing and GIS. GIS-Zeitschrift fur Geoinformationssysteme 14(6): 12-17.
  4. Cho, H.K., Lee, W.K. and Lee, S.H. 2003. Mapping of Vegetation Cover using Segment Based Classification of IKONOS Imagery. Korean Journal of Ecology 26(2): 75- 81. https://doi.org/10.5141/JEFB.2003.26.2.075
  5. Chubey, M.S., Franklin, S.E. and Wulder, M.A. 2006. Objectbased Analysis of Ikonos-2 Imagery for Extraction of Forest Inventory Parameters. Photogrammertic Engineering and Remote Sensing 72(4): 383-394.
  6. Cognalton, R.G. and Green, K. 1999. Assessing the Accuracy of Remotely Sensed Data:Principles and Practices, Boca Raton, FL : Lewis Publishers. 137p.
  7. Digital Globe. 2004. DigitalGlobe. www.digitalglobe.com.
  8. Franklin, S.E., Maudie, A.J. and Larvigne, M.B. 2001. Using spatial co-occurrence texture to increase forest structure and species composition classification accuracy. Photogrammetric Engineering and Remote Sensing 67: 849-855.
  9. Harvey, K.R. and Hill, G.J.E. 2001. Vegetation mapping of a tropical freshwater swamp in the Northern Territory, Australia: a comparison of aerial photography, Landsat TM and SPOT satellite imagery. International Journal of Remote Sensing 22: 2911-25.
  10. Hayes, D.J. and Sader, S.A. 2001. Comparison of changedetection techniques for monitoring tropical forest clearing and vegetation regrowth in a time series. Photogrammetric Engineering and Remote Sensing 67: 1067-1075.
  11. Jensen, J.R. 2004. Introductory Digital Image Processing : A Remote Sensing perspective. 3rd Edition, Prentice Hall. NJ. USA.
  12. Landis, J. and Koch, G. 1977. The Measurement of Observer Agreement for Categorical Data. Biometrics 33: 159-174. https://doi.org/10.2307/2529310
  13. Leckie, D.G. and Gillis, M.D. 1995. Forest Inventory in Canada with emphasis on map production. The Forestry Chronicle 71(1): 74-88.
  14. McRoberts, R. and Tomppo, E. 2007. Remote sensing support for national forest inventories. Remote Sensing of Environment 110(4): 412-419. https://doi.org/10.1016/j.rse.2006.09.034
  15. Smith, A. 2011. Image segmentation scale parameter optimization and land cover classification using the Random Forest algorithm. Journal of Spatial Science 55(1): 69-79.
  16. Wang, L., Sousa, W.P., Gong, P. and Biging, G.S. 2004. Comparison of IKONOS and QuickBird images for mapping mangrove species on the Caribbean coast of Panama. Remote Sensing of Environment 91(3-4): 432-440. https://doi.org/10.1016/j.rse.2004.04.005
  17. Yarbrough, L.D., Greg, E. and Joel, S.K. 2005. Quickbird 2 Tasseled Cap Transform Coefficients: A Comparison of Derivation Methods. Proceedings of Pecora 16 "Global Priorities in Land Remote Sensing". October 23-27, 2005, Sioux Falls, South Dakot.
  18. Yim, J.S., Kleinn, C., Cho, H.K. and Shin, M.Y. 2010. Integration of Digital Satellite Data and Forest Inventory Data for Forest Cover Mapping in Korea. Forest Science and Technology 6(2): 87-96. https://doi.org/10.1080/21580103.2010.9671976