软件学报
軟件學報
연건학보
JOURNAL OF SOFTWARE
2014年
7期
1570-1582
,共13页
唐利明%王洪珂%陈照辉%黄大荣
唐利明%王洪珂%陳照輝%黃大榮
당리명%왕홍가%진조휘%황대영
变分水平集%图像聚类%图像分割%模糊聚类%聚类中心
變分水平集%圖像聚類%圖像分割%模糊聚類%聚類中心
변분수평집%도상취류%도상분할%모호취류%취류중심
variational level set%image clustering%image segmentation%fuzzy clustering%clustering center
结合变分水平集方法和模糊聚类,提出了一个基于变分水平集的图像聚类分割模型。该模型引入了一个基于图像局部信息的外部模糊聚类能量和一个新的关于零水平集的正则化能量,使得该模型对噪声图像的聚类分割更具鲁棒性。通过在能量泛函中加入一个内部约束能量约束水平集函数为符号距离函数,可以使水平集演化过程无需重新初始化。进一步提出了一种变分形式的聚类中心更新方法,实现了半监督的图像聚类分割。实验中采用不同类型的图像与 FCM 聚类模型、CV 模型、Samson 模型进行了对比实验,实验结果显示,该模型能够克服图像中噪声的影响,取得较满意的聚类分割效果。
結閤變分水平集方法和模糊聚類,提齣瞭一箇基于變分水平集的圖像聚類分割模型。該模型引入瞭一箇基于圖像跼部信息的外部模糊聚類能量和一箇新的關于零水平集的正則化能量,使得該模型對譟聲圖像的聚類分割更具魯棒性。通過在能量汎函中加入一箇內部約束能量約束水平集函數為符號距離函數,可以使水平集縯化過程無需重新初始化。進一步提齣瞭一種變分形式的聚類中心更新方法,實現瞭半鑑督的圖像聚類分割。實驗中採用不同類型的圖像與 FCM 聚類模型、CV 模型、Samson 模型進行瞭對比實驗,實驗結果顯示,該模型能夠剋服圖像中譟聲的影響,取得較滿意的聚類分割效果。
결합변분수평집방법화모호취류,제출료일개기우변분수평집적도상취류분할모형。해모형인입료일개기우도상국부신식적외부모호취류능량화일개신적관우령수평집적정칙화능량,사득해모형대조성도상적취류분할경구로봉성。통과재능량범함중가입일개내부약속능량약속수평집함수위부호거리함수,가이사수평집연화과정무수중신초시화。진일보제출료일충변분형식적취류중심경신방법,실현료반감독적도상취류분할。실험중채용불동류형적도상여 FCM 취류모형、CV 모형、Samson 모형진행료대비실험,실험결과현시,해모형능구극복도상중조성적영향,취득교만의적취류분할효과。
An image clustering segmentation model combined with variational level set and fuzzy clustering is proposed in this paper. An external fuzzy clustering energy based on the local image information and a new regularization energy with respect to the zero level set are introduced in the energy functional, which makes the proposed model robust in noisy image segmentation. An internal energy that forces the level set function to be close to a signed distance function is introduced in the energy functional, which can completely eliminate the need of the expensive periodical re-initialization procedure for level set function during its evolution. Furthermore, this paper proposes a variational formulation to update the cluster centers in the procedure of clustering, which realizes the semi-supervised clustering segmentation. The experimental results show that the proposed model, compared with the FCM clustering model, CV model and Samson model, can reduce the influence of noise and get better segmentation results for different kinds of images.