한국지능시스템학회:학술대회논문집 (Proceedings of the Korean Institute of Intelligent Systems Conference)
- 한국퍼지및지능시스템학회 2002년도 추계학술대회 및 정기총회
- /
- Pages.333-336
- /
- 2002
Adaptive Threshold Determination Using Global and local Fuzzy Measures
- Jin, Mun-Gwang (Dept. of Information Engineering Myong Ji University) ;
- Woo, Dong-Min (Dept. of Information Engineering Myong Ji University) ;
- Lee, Kyu-Wong (Div. of Computer & Communications Engineering, Daejeon University)
- 발행 : 2002.12.01
초록
This paper presents a new image segmentation method using fuzzy measures which reflect the local property of an image as well as the global property of an image An image is globally segmented into the crisp region and the ambiguous region in terms of the Index of fuzziness measured over all pixels of an image. The ambiguous region is luther partitioned into background and object in terms of the index of fuzziness computed over the set of neighboring pixels reflecting the local property most. From the experimental results, this method shows the effective ambiguity handling capability in segmenting an image.
키워드