DOI QR코드

DOI QR Code

혼합화소를 이용한 IKONOS 영상의 감독분류정확도 평가

Accuracy Evaluation of Supervised Classification about IKONOS Imagery using Mixed Pixels

  • Lee, Jong-Sin (Dept. of Civil Engineering, Chungnam National University) ;
  • Kim, Min-Gyu (Dept. of Civil Engineering, Chungnam National University) ;
  • Park, Joon-Kyu (Dept. of Civil Engineering, Seoil College)
  • 투고 : 2012.03.29
  • 심사 : 2012.06.07
  • 발행 : 2012.06.30

초록

위성영상을 이용한 감독분류에서 훈련집단의 선택은 분류정확도에 많은 영향을 미친다. 일반적으로 훈련집단의 특징이 명확한 순수화소를 선택할 경우 전체 정확도가 높은 반면, 저해상도 영상이거나 식별이 불분명하여 혼합화소를 선택하면 정확도는 저하된다. 그러나 실제 영상분류를 수행할 때 순수화소만을 훈련집단으로 선택하는 것은 매우 어렵다. 이에 본 연구에서는 혼합화소를 훈련집단으로 선택하였을 경우 적합한 분류기법을 제시하고자 하였다. 이를 위해 소수의 순수화소를 훈련집단으로 선정하여 분류정확도를 산출하고 같은 수의 혼합화소를 이용한 분류결과와 정확도를 비교하였다. 연구 결과, 혼합화소를 사용한 분류기법들 중 SVM의 정확도가 가장 높았으며, 순수화소를 이용한 분류결과와도 가장 작은 차이를 보였다. 따라서 훈련집단으로 혼합화소를 선택할 가능성이 높은 건물 및 녹지혼합지역에서는 SVM을 이용한 영상분류가 가장 적합할 것으로 판단된다.

Selection of training set influences the classification accuracy in supervised classification using satellite imagery. Generally, if pure pixels which character of training set is clear were selected, whole accuracy is high while if mixed pixels were selected, accuracy is decreased because of low-resolution imagery or unclear distinguishment. However, it is too difficult to choose the pure pixels as training set actually. Accordingly, this study should be suggested the suitable classification method in case of mixed pixels choice. To achieve this, a few pure pixels were chosen as training set and classification accuracy was calculated which was compared with classification result using an equal number of mixed pixels. As a result, accuracy of SVM was the highest among the classification method using mixed pixels and it was a relatively small difference with the result of classification using pure pixels. Therefore, imagery classification using SVM is most suitable in the mixed area of construction and green because it is high possibility to choose mixed pixels as training set.

키워드

참고문헌

  1. H. C. Yun, J. S. Lee, K. Y. Jung, J. K. Park, "The Evaluation of Classification Precision about Satellite Imagery by the Selection of Training Set and Classification Method", Proceedings of Conference on Geo-Spatial Information 2011 Fall, pp.286-287, 2011.
  2. K. E. Kim, "A Modified Iterative N-FINDR Algorithm for Fully Automatic Extraction of Endmembers from Hyperspectral Imagery", Korean Journal of Remote Sensing, Vol.27, No.5, pp.565-572, 2011. https://doi.org/10.7780/kjrs.2011.27.5.565
  3. Y. W. Seok, "A Study on the Optimum Band Selection for Supervised Classification Using Mixed-Pixels and Canonical Correlation Analysis", Thesis of Master, SunKyunKwan Univ., 2004.
  4. H. Y. Lim, H. S. Kim, J. S. Choi, S. H. Song, "A Study on Efficient Topography Classification of High Resolution Satellite Image", Journal of The Korean Society for GeoSpatial Information System, Vol.13, No.3, pp.33-40, 2005.
  5. J. M. Kang, H. C. Yun, J. B. Kim, J. S. Lee, "The Evaluation of Classification Precision about Satellite Imagery by the Selection Method of Training Set", Proceedings of Korean Society of Civil Engineers 2010 Annual Conference, pp.1082-1084, 2010.
  6. G. M. Foody, A. Mathur, "The use of small training sets containing mixed pixels for accurate hard image classification: Training on mixed spectral responses for classification by a SVM", Remote Sensing of Environment, Vol.103, pp.179-189, 2006. https://doi.org/10.1016/j.rse.2006.04.001
  7. M. Chi, R. Feng, L. Bruzzone, "Classification of hyperspectral remote-sensing data with primal SVM for small-sized training dataset problem", Advances in Space Research, Vol.41, pp.1793-1799, 2008. https://doi.org/10.1016/j.asr.2008.02.012
  8. K. T. Hur, "A Study to Compare Classification Using the IKONOS Satellite Imagery", Thesis of Master, Chonbuk National Univ., 2012.