QQ客服 官方微博 反馈留言

	基于特征聚类的舌下络脉自动提取方法
	孙丹萍 ; 吴佳 ; 张永红 ; 白净 ; 翁维良 ; 吴煜 ; SUN Dan-Ping ; WU Jia ; ZHANG Yong-Hong ; BAI Jing ; WENG Wei-Liang ; WU Yu
	2010-05-13 ; 2010-05-13
关键词	中医舌诊 舌下络脉 自动提取 K-均值聚类算法 L*a*b*色彩空间 Chinese Medicine sublingual venae automatic extraction K-means clustering L*a*b* color space R319 R241.25
其他题名	Automatic Sublingual Venae Extraction Method Based on Clustering
中文摘要	舌下络脉诊是中医舌诊中十分重要的一种手段,本研究提出了一种基于特征聚类分析的舌下络脉自动提取方法,该算法在L*a*b*色彩空间下,利用K-均值聚类算法对舌下区域按颜色进行聚类,根据聚类中心的位置,确定舌下络脉所在的聚类,并计算出舌下络脉的特征参数,对每一例舌下数据给出6个参数,表征络脉的长度,宽度以及颜色。应用此算法对北京东直门医院886例肝病病人的舌下图像数据进行分析,经中医医生判断,成功提取舌下络脉839例,成功率达96.88%。; Sublingual venae consultation is an important means of traditional tongue diagnosis in Chinese Medicine.In this paper,an automatic sublingual venae extraction method based on K-means clustering was proposed.The sublingual pictures were first converted from RGB color space to L*a*b* color space,then processed using K-means clustering algorithm to cluster sublingual areas by colors.According to the cluster centers,we finally decided which cluster the venae are in.At the same time,feature parameters related to sublingual venae were calculated.There were 6 parameters for each venae picture,describing the length,width and color of the detected venae.This method was verified on 886 sublingual pictures of liver disease patients in Dongzhimen Hospital,Beijing.According to the diagnosis of doctors,96.88% of all the 886 pictures,area were extracted correctly. The proposed method shows its potential to be applied in clinical analysis.
语种	中文 ; 中文
内容类型	期刊论文
源URL	[http://hdl.handle.net/123456789/32690]
专题	清华大学
推荐引用方式 GB/T 7714	孙丹萍,吴佳,张永红,等. 基于特征聚类的舌下络脉自动提取方法[J],2010, 2010.
APA	孙丹萍.,吴佳.,张永红.,白净.,翁维良.,...&WU Yu.(2010).基于特征聚类的舌下络脉自动提取方法..
MLA	孙丹萍,et al."基于特征聚类的舌下络脉自动提取方法".(2010).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。