Abstract
The classical approach of using minimum cut criterion for clustering is often ineffective due to the existence of outliers in the data. This paper presents a novel normalized graph sampling algorithm for clustering that improves the solution of clustering via the incorporation of a priori constraint in a stochastic graph sampling procedure. The quality of the proposed algorithm is empirically evaluated on two synthetic datasets and a color medical image database.
Original language | English |
---|---|
Pages (from-to) | 819-822 |
Number of pages | 4 |
Journal | Proceedings - International Conference on Pattern Recognition |
Volume | 16 |
Issue number | 3 |
Publication status | Published - 2002 |
Scopus Subject Areas
- Computer Vision and Pattern Recognition