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

The study on Decision Tree method to improve land cover classification accuracy of Hyperspectral Image

초분광영상의 토지피복분류 정확도 향상을 위한 Decision Tree 기법 연구

  • SEO, Jin-Jae (Korea Land and Geospatial Informatix Corporation) ;
  • CHO, Gi-Sung (Dept. of Civil Engineering, Chonbuk National University) ;
  • SONG, Jang-Ki (Dept. of Civil Engineering, Chonbuk National University)
  • Received : 2018.09.13
  • Accepted : 2018.09.30
  • Published : 2018.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Hyperspectral image is more increasing spectral resolution that Multi-spectral image. Because of that, each pixel of the hyperspectral image includes much more information and it is considered the most appropriate technic for land cover classification. but recent research of hyperspectral image is stayed land cover classification of general level. therefore we classified land cover of detail level using ED, SAM, SSS method and made Decision Tree from result of that. As a result, the overall accuracy of general level was improved by 1.68% and the overall accuracy of detail level was improved by 5.56%.

초분광영상(Hyperspectral Image)은 다중분광영상에 비해 각 픽셀이 가지는 정보량이 많아 다양한 토지피복을 분류하는데 있어 가장 적합한 영상으로 평가 받고 있다. 하지만 최근의 초분광영상의 연구는 대분류에 해당하는 연구에 그치고 있다. 이에 본 연구에서는 다양한 토지피복분류에 대한 연구를 수행하기 위해 기존의 분석기법인 ED, SAM, SSS 기법을 토대로 Decision Tree를 구성하는 연구를 수행하였다. 그 결과, 대분류의 전체정확도는 1.68%, 세분류 전체정확도는 5.56%가 향상되는 결과를 얻을 수 있었다.

Keywords

References

  1. Adams J.B. and A.R. Gillespie. 2006. Remote sensing of landscapes with spectral images - physical modeling approach. United Kingdom at the University Press Cambridge.
  2. Carvalho O.A. and P.R. Meneses. 2000. Spectral Correlation Mapper(SCM) : AN Improvement on the Spectral Angle Mapper(SAM). 9th JPL Airborne Earth Science Workshop. JPL publication
  3. Kim, S.H., G.S. Lee, J.L. Ma and M.J. Kook. 2005. Current satatus of hyperspectral remote sensing:principle, data processing technique, Journal of Korea Society of Remote Sensing, 21(4):341-369.
  4. Kim, Y.H., D.S. Kim, Y.I. Kim and K.Y. Yu. 2007. The application of the spectral similarity scale algorithm and expectation-maximization for unsupervised change detection using hyperspectral image, Proc. 2007 Conference of Korean Society for Geo-Spatioal Information System, pp.139-144.
  5. Lee, K.C. 2014. Land cover classification using airborne hyperspectral imagery, Master's thesis, Chonbuk National University.
  6. Park, J.S. 2016. Study on the spectral library establishing and availability for analysis of hyperspectral image, Master's thesis, Chonbuk National University.
  7. Park, M.H. 2009. A study on feature selection and feature extraction for hypersepctral image classification using canonical correlation classifier, Journal of Korean Society of Civil Engineers, 24(4):341-369.
  8. Seo, J.J. 2017. The study on land cover classification of hyperspectral image using Decision Tree Method, Master's thesis, Chonbuk National University.
  9. Shin, J.I., S.H. Kim and G.S. Lee. Research articles : design and construction of spectral library for the Korean peninsular, Journal of Korea Society of Remote Sensing 26(5):465-475.
  10. Sweet J.N. 2003. The Spectral Similarity Scale and its Application to the Classification of Hyperspectral Remote Sensing Data. IEEE Workshop on Advances in techniques for Analysis of Remotely Sensed data. pp.92-99.