| 注册
首页|期刊导航|西安工程大学学报|纬编针织物疵点的图像分割方法

纬编针织物疵点的图像分割方法

孙尧 龙海如

西安工程大学学报2009,Vol.23Issue(2):230-234,5.
西安工程大学学报2009,Vol.23Issue(2):230-234,5.

纬编针织物疵点的图像分割方法

Method of segment weft knitted fabric defects image

孙尧 1龙海如1

作者信息

  • 1. 东华大学,纺织学院,上海,201620
  • 折叠

摘要

Abstract

The research of defect detection is mainly depended on the segmentation between defect image and background image, and the defect information such as energy, entropy, ete are calculated. Two methods of image segmentation are compared, which are Otsu's method and Cellular Neural Network (CNN). And a new method of enhancing image is introduced. The interlock fabric with the defects of hole, course mark and dropped stitch are used for experiment materials. The experiment proved the performance of CNN method is better than Otsu, and the method is well performed for promoting segment effect.

关键词

疵点检测/Otsu法/细胞神经网络/图像增强

Key words

defect detection/Otsu's method/CNN/image enhancing

分类

轻工纺织

引用本文复制引用

孙尧,龙海如..纬编针织物疵点的图像分割方法[J].西安工程大学学报,2009,23(2):230-234,5.

西安工程大学学报

OACSTPCD

1674-649X

访问量0
|
下载量0
段落导航相关论文