首页出版物导航 用户协会 博客
 精确匹配 高级检索 检索历史
河北工业大学
详细信息

基于图像处理的织物覆盖系数检测.

Title: 基于图像处理的织物覆盖系数检测.
Alternate Title: Inspection on the cover factors of fabric based on image processing.
Language: Chinese
Authors: 陈阳1
陈霞1,2 chenxia@dhu.edu.cn
汪军1
Source: Journal of Xi'an Polytechnic University. 2016, Vol. 30 Issue 1, p16-20. 5p.
Document Type: Article
Author-Supplied Keywords: Canny edge detection ; cover factor ; image processing ; local threshold ; staple yarn ; woven fabric ; 图像处理 ; 局部阈值 ; 机织物 ; 短纤纱 ; 覆盖系数 ; Canny 边缘检测 ; Language of Keywords: English; Chinese
Abstract (English): To reduce the influence of hairiness interference and background unevenness on staple fabric image, the local threshold segmentation and Canny edge analysis were applied to detect the cover factors of fabric. Firstly, the fabric image was divided into a series of local images by yarn centerlines. Using Canny edge detection to obtain effective local threshold, these local images were transformed into binary images. After morphological processing, these local binary images were merged into a complete binary image and cover factor was calculated. Finally, the actual cover factor of three plain fabric samples was compared with theoretical values. The correlation coefficient between the actual and theoretical values was 0.99. The result shows that the method in this paper can detect the cover factors of staple yarn fabric effectively. [ABSTRACT FROM AUTHOR]
Abstract (Chinese): 针对短纤纱织物图像中毛羽的干扰和图像背景不匀,将 Canny 边缘分析和局部阈值分割法结合用于检测织物的覆盖系数.首先根据纱线中心线将织物图像划分成许多局部图像,利用Canny 算子检测局部图像中孔隙的边缘,确定有效局部阈值并转换成二值图像.然后将局部二值图像形态学处理后融合得到完整的织物二值图像,计算覆盖系数.最后对3种自织的平纹棉织物的覆盖系数进行检测,并与理论计算的覆盖系数比较,两者相关系数达0.99,说明此方法可以较好地检测短纤纱织物的覆盖系数. [ABSTRACT FROM AUTHOR]
(Copyright applies to all Abstracts.)
Author Affiliations: 1东华大学 纺织学院,上海,201620 东华大学 纺织学院,上海 201620 ; 2东华大学 产业用纺织品教育部工程研究中心,上海 201620
ISSN: 1674-649X (Sherpa/RoMEO, JCR)
PageCount: 16-21
volume: 30
issue: 1
issn: 1674649X
pubdate: 2016
DOI: https://doi.org/10.13338/j.issn.1674-649x.2016.01.003
JCR分区:
ESI学科:
最新一年影响因子:
当年影响因子:
Custom Links: Full Text Finder