几种织物疵点检测算法的设计与比较  

Design and Comparison of Several Fabric Defect Detection Algorithms

作　　者：徐园园[1] 刘伟斌[1] 郑力新[2] 

机构地区：[1]华侨大学信息工程与科学学院,福建厦门361021 [2]华侨大学工学院,福建泉州362021

出　　处：《控制工程》2016年第5期777-782,共6页Control Engineering of China

基　　金：国家自然科学基金项目(F030306);福建省泉州市科技计划项目(2015Z124);福建省科技厅项目(2013H2002)

摘　　要：在研究相关文献内容的基础上,选取目前织物疵点自动检测中3种常用的特征提取算法,基于Hough变换和Gabor滤波的检测算法、基于频域筛状滤波器的检测算法和基于灰度共生矩阵法的检测算法,简要介绍了3种检测算法的设计原理和实现步骤。基于相同的硬件、软件环境,分别使用3种检测算法对几种常见的织物疵点进行测试。最后通过对时效性、检测效果等进行比较分析,3种检测算法对于纹理性、非纹理性等不同的疵点具有不同的优缺点。Based on the research of related literatures, this paper analyzes three commonly used feature extraction algorithms in automatic fabric defect detection in the current research. The detection algorithms include Hough transformation and Gabor filter, the frequency domain sieve-like filter and gray level co-occurrence matrix. Firstly, the design principles and the implementation steps of the three algorithms are briefly introduced. Then, based on the same hardware and software environment, the three detection algorithms are used for several common defects respectively. Finally, this paper analyzes the comparative results of the three algorithms in terms of timeliness and detection effects. Three algorithms have different advantages and disadvantages for texture, non-textured of different defects.

关 键 词：织物疵点 特征提取 检测算法 比较 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]