Korean Journal of Remote Sensing (대한원격탐사학회지)

Korean Society of Remote Sensing (대한원격탐사학회)
권호 표지
DOI QR코드

DOI QR Code

Fuzzy Training Based on Segmentation Using Spatial Region Growing

  • Lee Sang-Hoon (Department of Industrial Engineering, Kyungwon University)
  • Published : 2004.10.01
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes an approach to unsupervisedly estimate the number of classes and the parameters of defining the classes in order to train the classifier. In the proposed method, the image is segmented using a spatial region growing based on hierarchical clustering, and fuzzy training is then employed to find the sample classes that well represent the ground truth. For cluster validation, this approach iteratively estimates the class-parameters in the fuzzy training for the sample classes and continuously computes the log-likelihood ratio of two consecutive class-numbers. The maximum ratio rule is applied to determine the optimal number of classes. The experimental results show that the new scheme proposed in this study could be used to select the regions with different characteristics existed on the scene of observed image as an alternative of field survey that is so expensive.

Keywords

References

  1. Anderberg， M. R, 1973. Cluster Analysis for Application, Academic Press, NewYork
  2. Lee, S-H., 2001. Unsupervised image classification using spatial region growing segmentation and hierarchical clustering, 17(1):57-70 (in Korean)
  3. Lee, S-H., 2004. Unsupervised image classification using region-growing segmentation based on CN-chain, 20(3): 235-248 (inKorean)
  4. Liang, Z, R. J. Jaszczak, and R. E. Coleman, 1992. Parameter estimation of finite mixture using the EM algorithm and information criteria with application to medical image processing, IEEE Trans. Nucl.Sci., 39: 1126-1133 https://doi.org/10.1109/23.159772
  5. Duda, R. O., P. E. Hart and D. G. Stork, 2001. Pattern Classification, Wiley Interscience, NewYork