摘要
为实现自动检测织物图案周期和印花织物质量的有效评估,运用距离匹配函数计算印花织物周期,首先计算印花织物图案行和列的一维距离匹配函数,然后对织物所有行和列的一维距离匹配函数求和作为行和列的二维距离匹配函数。当距离匹配函数中的周期变量为织物图案周期时,印花织物距离匹配函数二次前差分得到最大值。结果表明,与传统的自相关函数检测织物周期相比,该算法检测织物周期准确且可检测织物各种图案的周期。
In order to realize automatic detection of design cycle of fabrics and quality evaluation of printed fabrics effectively,in this paper,cycle of printed fabrics is determined via calculating distance matching function. One-dimensional distance matching function of each row and each column of patterned fabrics is calculated firstly based on cyclical pattern fabrics,and the sums of two-dimensional distance matching function on all rows and columns are acted as results of two-dimensional distance matching function on row and column. When the variable in two-dimensional distance matching function is the cycle of patterned fabrics,second forward difference of two-dimensional distance matching function reaches the maximum. The experimental results show two-dimensional distance matching function of inspecting cycles of all kinds of fabrics are efficient and accurate. Compared with conventional autocorrelation function for patterned cycle detection,the proposed method in this paper can achieve more accurate cycle detection of patterned fabrics,and cycles of various patterned fabrics can be inspected,thus a foundation is provided for real-time operation in industry.
出处
《纺织学报》
EI
CAS
CSCD
北大核心
2015年第12期98-103,共6页
Journal of Textile Research
基金
国家自然科学基金资助项目(61301276)
西安工程大学博士科研启动基金资助项目(BS1416)
西安工程大学学科建设经费资助项目(107090811)
关键词
印花织物
图案周期
距离匹配函数
前差分
printed fabrics
cycles of patterned
distance matching function
forward difference