摘要
根据小波在奇异信号分解中的特点,提出了一种基于小波分解的疵点检测新方法.首先根据织物纹理特点,确定小波函数.其次对被检测图像进行小波变换,获得分解后的子图;根据织物纹理组织单元,把高频子图分割成若干子窗口,统计子窗口的能量标准差与均值加权求和作为提取的特征.最后通过测试图像子窗口特征与标准子窗口特征相比较,判断疵点是否存在.实验表明,该检测方法是有效的,检测正确率达到90%以上.
According to the characteristics of wavelet decomposition in singular signals, an innovative method for defect detection based on wavelet decomposition is put forward. Firstly, according to fabric texture characteristics, wavelet functions are identified. Secondly, sub -images are acquired through the fabric image wavelet decomposition. High frequency sub - images are segmented into many sub - windows, in which stand deviation and average weighted sum are taken as characteristics. Lastly, the characteristics are compared with normal sub - window＇s characteristics to determine whether there is a defect. The experimental result confirms that the proposed method is feasible in rapid defect detection, and the detection average accuracy is over 90%.
出处
《昆明理工大学学报（理工版）》
北大核心
2009年第1期48-51,103,共5页
Journal of Kunming University of Science and Technology(Natural Science Edition)
基金
西安市科技攻关资助项目(项目编号:GG04039)
关键词
小波分解
特征提取
疵点检测
织物
wavelet decomposition
characteristic extraction
defect detection
fabric
