DOI QR코드

DOI QR Code

항공 하이퍼스펙트럴 영상의 PCA기법 적용을 통한 토지 피복 분류 정확도 개선 방안에 관한 연구

A Study on the Improvement classification accuracy of Land Cover using the Aerial hyperspectral image with PCA

  • 최병길 (인천대학교 도시과학대학 건설환경공학과) ;
  • 나영우 (인천대학교 산학협력) ;
  • 김승현 (선영엔지니어링 기술영업부) ;
  • 이정일 (인천대학교 일반대학원 건설환경공학과)
  • Choi, Byoung Gil (Dept. of Civil and Environmental Engineering, Incheon National University) ;
  • Na, Young Woo (Hub-Industry-Academic-Cooperation, Incheon National University) ;
  • Kim, Seung Hyun (Sunyoung ENG) ;
  • Lee, Jung Il (Dept. of Civil and Environmental Engineering, Incheon National University)
  • 투고 : 2014.02.13
  • 심사 : 2014.03.18
  • 발행 : 2014.03.31

초록

본 연구에서는 항공 하이퍼스펙트럴 영상에 대해 PCA를 적용하여 토지 이용 및 피복 분류 판독의 가독성을 향상시키기 위하여 고유성분이 높은 밴드를 선별적으로 조합하여 5개 유형의 PCA영상을 제작하였다. 유형별 영상은 SAM감독 분류 기법을 적용하여 영상분류를 시행하고 정확도를 평가한 결과 PCA변환 시 고유성분 포함율은 PCA변환 영상의 첫 번째 밴드에 해당하는 영상이 76.74%의 성분을 포함하며, PCA변환 영상의 두 번째 누적 밴드에 해당하는 영상이 98.40%로 대부분의 성분자료가 두 번째 영상까지에 담긴 것을 알 수 있었다. 유형별 영상의 정량적 분류정확도 평가는 전체정확도, 생산자 및 사용자 정확도를 분석한 결과 유사한 패턴을 가지며, 특이한 사항은 정성적인 분류정확도 평가는 PCA변환 영상의 네 번째 밴드이상이 포함되어야 정확도가 확보되는 것으로 판단되나 정량적인 분류 정확도 평가에서는 PCA변환 영상의 두 번째 밴드까지를 포함하는 영상이 가장 높은 정확도를 나타내는 것을 알 수 있었다.

The researcher of this study applied PCA on aerial hyper-spectral sensor and selectively combined bands which contain high amount of information, creating five types of PCA images. By applying Spectral Angle Mapping-supervised classification technique on each type of image, classification process was carried out and accuracy was evaluated. The test result showed that the amount of information contained in the first band of PCA-transformation image was 76.74% and the second accumulated band contained 98.40%, suggesting that most of information were contained in the first and the second PCA components. Quantitative classification accuracy evaluation of each type of image showed that total accuracy, producer's accuracy and user's accuracy had similar patterns. What drew the researcher's attention was the fact that the first and the second bands of the PCA-transformation image had the highest accuracy according to the classification accuracy although it was believed that more than four bands of PCA-transformation image should be contained in order to secure accuracy when doing the qualitative classification accuracy.

키워드

참고문헌

  1. Cho, D. Y. and Kim, E. M., 2010, Extraction of spatial information of tree using LIDAR data in urban area, Journal of The Korean Society for Geospatial Information System, Vol. 18 No. 4, pp.11-20.
  2. Choi, J. W., Kim, D. S., Lee, B. K., Kim, Y. I. and Yu, K. Y., 2006, Hyperspectral image fusion algorithm based on two-stage spectral unmixing method, Korean Journal of Remote Sensing, Vol. 22 No.4, pp.295-304. https://doi.org/10.7780/kjrs.2006.22.4.295
  3. Du, Q., 2012, A new sequential algorithm for hyperspectral endmember extraction, IEEE Geoscience and Remote Sensing Letters, Vol.9 No.4, pp.695-699. https://doi.org/10.1109/LGRS.2011.2178815
  4. Han, D. Y., Cho, Y. W., Kim, Y. I. and Lee, Y. W., 2003, Feature selection for image classification of hyperion data, Korean Journal of Remote Sensing, Vol. 19 No.2, pp.171-179. https://doi.org/10.7780/kjrs.2003.19.2.18
  5. Han, Y. K., Byun, Y. K, and Kim, Y. I., 2012, Automatic estimation of geometric translations between high-resolution optical and SAR images, Journal of the Korean Society for Geospatial Information System, Vol. 20 No. 3, pp.41-48. https://doi.org/10.7319/kogsis.2012.20.3.041
  6. Lee, C. K., Kim, B. O. and Kim, N. Y., 2007, Investigation of shoreline change by photogrammetric method, Journal of Korean Society of Geospatial Information System, Vol. 15 No. 2, pp.15-23.
  7. Liangrocapart, S. and Petrou, M., 1998, Mixed pixels classification, Proc. SPIE 3500, Image and Signal Processing for Remote Sensing IV, pp.72-83.
  8. Mars, J. C., Garrity, C. P., Houseknecht, D. W., Amoroso, L. and Mears, D. C., 2005, Digital-elevation and surface-classification maps of the Fish Creek Area, Harrison Bay Quadrangle, Northern Alaska, U.S. Geological Survey.
  9. Park, M. H., 2006, Usefulness of canonical correlation classification technique in hyper-spectral image classification, Korea Society of Civil Engineers, Vol. 26 No. 5D, pp.885-894.
  10. Shin, J. I., Kim, S. H., Yoon, J. S., Kim. T. G. and Lee, K. S., 2006, Spectral mixture analysis using hyperspectral image for hydrological land cover classification in urban area, Korean Journal of Remote Sensing, Vol. 22 No. 6, pp.565-574. https://doi.org/10.7780/kjrs.2006.22.6.565

피인용 문헌

  1. 장성 백양사 소요대사탑의 비파괴 훼손도 진단과 입지환경 검토 vol.49, pp.4, 2014, https://doi.org/10.22755/kjchs.2016.49.4.52