摘要
引入二维小波分析方法,分析针织物的外观平整度,提取了较为精细的针织物图像信息.图像首先经过中值滤波,再利用小波变换分解并从中提取高频信息,然后结合4种表征织物折皱的特征参数,计算不同平整度等级模板的特征值.并以此为依据,运用最小距离判别法,确定16种不同针织物类型的平整度等级.为了定量描述评定结果,通过计算客观评定与主观评定结果的相关系数,验证该方法的可行性.
Two-dimensional wavelet transform was used to analyse knitted fabric smoothness in order to acquire the finer image information. Firstly, knitted fabric image was filtered through median filter, and decomposed by wavelet transform; meanwhile, high frequency information was extracted. Secondly, four kinds of wrinkle feature parameters were applied to calculate the feature values of fabric smoothness with different smoothness templates. Finally, smoothness grade of sixteen different types of knitted fabrics can be evaluated according to this result through using minimum distance classification. For describing the assessment result quantitatively, the correlation coefficient was calculated between objective assessment and subjective assessment to validate the feasibility of this method.
出处
《东华大学学报（自然科学版）》
CAS
CSCD
北大核心
2008年第6期692-695,699,共5页
Journal of Donghua University(Natural Science)
关键词
十织物平整度
等级评定
小波分析
特征值提取
knitted fabric smoothness
grade assessment
wavelet analysis
feature values extraction
