Proceedings of the KSRS Conference (대한원격탐사학회:학술대회논문집)
- 2002.10a
- /
- Pages.54-58
- /
- 2002
- /
- 1226-9743(pISSN)
Edge preserving method using mean curvature diffusion in aerial imagery
- Ye, Chul-Soo (Spatial Imagery Information Research Team Spatial Information Technology Center, ETRI- Computer & Software Research Laboratory) ;
- Kim, Kyoung-Ok (Spatial Imagery Information Research Team Spatial Information Technology Center, ETRI- Computer & Software Research Laboratory) ;
- Yang, Young-Kyu (Spatial Imagery Information Research Team Spatial Information Technology Center, ETRI- Computer & Software Research Laboratory) ;
- Lee, Kwae-Hi (Dept. of Electronic Engineering, Sogang University)
- Published : 2002.10.01
Abstract
Mean curvature diffusion (MCD) is a selective smoothing technique that promotes smoothing within a region instead of smoothing across boundaries. By using mean curvature diffusion, noise is eliminated and edges are preserved. In this paper, we propose methods of automatic parameter selection and implementation for the MCD model coupled to min/max flow. The algorithm has been applied to high resolution aerial images and the results show that noise is eliminated and edges are preserved after removal of noise.