博碩士論文 975303013 完整後設資料紀錄

DC 欄位 語言
DC.contributor通訊工程學系在職專班zh_TW
DC.creator游健忠zh_TW
DC.creatorChien-Chung Yuen_US
dc.date.accessioned2011-7-30T07:39:07Z
dc.date.available2011-7-30T07:39:07Z
dc.date.issued2011
dc.identifier.urihttp://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=975303013
dc.contributor.department通訊工程學系在職專班zh_TW
DC.description國立中央大學zh_TW
DC.descriptionNational Central Universityen_US
dc.description.abstract目前在影像處理的領域中,圖像分割算是重要的一項影像處理。圖像分割可以透過色彩分群的方法達到圖像分割的目的。而透過圖像分割可以讓我們容易得到我們需要的部分,並將分離出來的部分加以處理。 模糊聚類演算法是屬於聚類演算法的一種,它是由K-Means改良而來的,它主要加上了模糊理論的概念,使得每一點的輸入向量以歸屬的程度來表現。但模糊聚類演算法本身有一些缺點,初始值的選擇會影像它分群的效果。所以我們利用群體智能粒子群優化演算法來找尋初始值,讓模糊聚類演算法來達到好的色彩分群效果。 所以我們本論文將結合模糊聚類演算法、粒子群演算法這二種來做為色彩分群,讓粒子群演算法做為初始值的搜尋的方法來改善模糊聚類演算法的收斂速度、分割的品質。 zh_TW
dc.description.abstractNowadays, image segmentation is an important technique in the image processing sector. We can easily extract the necessary parts from the entire image through this technique. Fuzzy C Means is a clustering algorithm coming from K-Means algorithm. The concept of fuzzy logic are applied to this method in which the performance on each point of the input vector has a degree of the belonging. The weakness of Fuzzy C Means is on the clustering effects when selecting the initial value of image. To achieve the good effect on image segmentation by using the Fuzzy C Means’ algorithm, we generate a method of intelligence particle swarm optimization algorithm to find the initial value. We used two algorithm methods in image segmentation technique which are Fuzzy C Means and particle swarm optimization algorithm. Both improved the convergence’s speed and the segmentation’s quality of Fuzzy C Means. en_US
DC.subject聚類演算法zh_TW
DC.subject粒子群演算法zh_TW
DC.subject群體智能zh_TW
DC.subject色彩分群zh_TW
DC.subject圖像分割zh_TW
DC.subjectImage segmentationen_US
DC.subjectFuzzy C Meansen_US
DC.subjectSwarm Intelligenceen_US
DC.subjectPSOen_US
DC.subjectParticle Swarm Optimizationen_US
DC.subjectClustering Algorithmsen_US
DC.title應用模糊聚類與粒子演算法之色彩分群研究zh_TW
dc.language.isozh-TWzh-TW
DC.titleFuzzy clustering and particle swarm optimization applied to color clustering algorithmen_US
DC.type博碩士論文zh_TW
DC.typethesisen_US
DC.publisherNational Central Universityen_US

若有論文相關問題,請聯絡國立中央大學圖書館推廣服務組 TEL:(03)422-7151轉57407,或E-mail聯絡  - 隱私權政策聲明