W-K-means algorithms spatial correction
碩士 === 國立新竹教育大學 === 人資處數學教育碩士班 === 97 === Fuzzy c-means clustering (FCM) with spatial constraints (FCM_S) is an effective algorithm suitable for image segmentation. Chen and Zhang (2004) propose robust kernelized versions KFCM_S, KFCM_S1 and KFCM_S2 by applying the kernel methods. Huang et al. (2005...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Online Access: | http://ndltd.ncl.edu.tw/handle/94060791525994397185 |
| Summary: | 碩士 === 國立新竹教育大學 === 人資處數學教育碩士班 === 97 === Fuzzy c-means clustering (FCM) with spatial constraints (FCM_S) is an effective algorithm suitable for image segmentation. Chen and Zhang (2004) propose robust kernelized versions KFCM_S, KFCM_S1 and KFCM_S2 by applying the kernel methods. Huang et al. (2005) proposes a W-k-means algorithm that can automatically calculate variable weights. In this paper we propose W-k-means algorithms with spatial correction, and we call them AWKM_S2 and RAWKM_S2. Some numerical and image experiments are performed to assess the performance of AWKM_S2 and RAWKM_S2 in comparison with KFCM_S2.Experimental results show that the proposed AWKM_S2 and RAWKM_S2 have better performance.
|
|---|