羊毛衫横纵密度的自动识别  

Automatic recognition of wale and course per unit length of woolen sweaters

作　　者：汪秀琛[1] 

机构地区：[1]中原工学院,河南郑州450007

出　　处：《纺织学报》2012年第4期60-63,共4页Journal of Textile Research

基　　金：河南省重大科技攻关项目(102102210064)

摘　　要：羊毛衫组织密度的识别目前还未有成熟算法。为此提出一套基于极值簇的密度自动识别方法。首先计算任意识别区域的实际尺寸,采用线性方法对图像进行增强处理;进一步提出极值簇融合算法,抽取图像中线圈圈柱及空隙部分产生的灰度峰值簇及谷值簇,并将同一位置的同类极值簇融合,以绘制能显现图像变化规律的横向及纵向灰度极值簇融合图;最后给出通过灰度极值簇融合图计算羊毛衫织物组织横纵密的公式。通过MatLab 7.0编程验证该算法,得出其对纯色平纹、罗纹组织识别准确率大于98%的结论。There is not a mature algorithm for density recognition of woolen sweater texture at present.Therefore,this paper sets forth an automatic recognition method based on extremum cluster.Firstly,actual size of any recognition area is calculated,and image is subjected to enhancement processing by linear method.Then a fusion algorithm of extremum is proposed.With this algorithm,gray peak cluster and valley cluster generated by loop column and void part in images are extracted,the identical extremum clusters in the same position are fused,and fusion graph of gray extremum cluster for image change rule is drawn.Finally,a calculation formula of wale and course per unit length of woolen sweater texture is given.The algorithm in this paper is verified by MatLab 7.0,and the results indicate that its recognition accuracy is more than 98% for jersey and rib texture.

关 键 词：羊毛衫 横密 纵密 极值簇 识别 

分 类 号：TS184.5[轻工技术与工程—纺织材料与纺织品设计]