摘要
针对帘子布疵点图像特征,提出了将小波变换和人工神经网络技术应用在帘子布疵点检测的方法。该方法是在融合图像灰度的基础上,经小波变换后再提取分解子图像的特征值,利用BP神经网络进行图像分类。实验结果表明,对帘子布常见疵点如油污、破洞、断经、断纬等能比较准确地识别。
A detection method for cord fabric defect was put forward according to its characeristics,using wavelet transform and neural network technology.The characteristics of the decomposed subimages were drawn based on the grayacale image.Then the characteristics information of defect were sent into BP neural network to classify defaults.The experimental results demonstrated that it could recognize four common fabric defects-weft-lacking,warp-lacking,oil stains and holes,and have advantages with high identification correctness.
出处
《纺织科技进展》
CAS
2008年第4期12-14,共3页
Progress in Textile Science & Technology
基金
河南省科技攻关项目(0721002210032)
关键词
机器视觉
小波变换
织物疵点
神经网络
图像处理
machine vision
wavelet transform
fabric defect
neural network
image processing
