织物纹理疵点快速检测方法  

Fabric texture defect for rapid detection method

作　　者：管声启[1,2] 石秀华[1] 

机构地区：[1]西北工业大学航海学院,陕西西安710072 [2]西安工程大学机电工程学院,陕西西安710048

出　　处：《计算机工程与设计》2009年第14期3433-3435,共3页Computer Engineering and Design

基　　金：西安市科技攻关基金项目(GG04039)

摘　　要：根据小波在奇异信号分解中的特点,提出了一种疵点快速检测新方法。首先通过对被检测图像进行小波提升分解,获得相应的子图。然后把高频子图分割成若干子窗口,统计子窗口能量的标准差与均值加权和作为提取的特征。最后把高频子图中的特征均值与正常子图特征均值相比较,判断疵点是否存在。实验结果表明,该检测方法能快速有效的检测疵点。According to wavelet decomposition characteristics in singular signals, an innovative method for rapid defect detection is put forward. Firstly, acquired images are decomposed by lifting scheme into sub-images. Secondly, high frequency sub-images are segmented into many sub-windows, in which standard deviation and average are calculated. Then they are weighted sum as characteristics. Lastly, the characteristic mean is used as basis of discrimination in every high frequency sub-image, which is compared with normal sub-image＇ s characteristic mean to determine whether there exists defect. The experimental result confirms that the proposed method can is feasible in rapid defect detection.

关 键 词：小波变换 提升分解 特征提取 特征均值 疵点检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]