基于独立成分分析的冷轧带钢表面缺陷识别

钢铁研究学报, 2011, 23(10): 63-67.

基于独立成分分析的冷轧带钢表面缺陷识别

1.1. 中国地质大学武汉机械与电子信息学院, 湖北武汉 430074 2. 中国地质大学武汉地球物理与空间信息学院, 湖北武汉 430074 3. 武汉钢铁集团公司, 湖北武汉 430083>

1.1. Faculty of Mechanical and Electronic Information, China University of Geosciences Wuhan, Wuhan 430074,Hubei, China 2. Institute of Geophysics and Geomatics, China University of Geosciences Wuhan, Wuhan430074, Hubei, China 3. Wuhan Iron and Steel Group Corporation, Wuhan 430083, Hubei, China>

基金项目: 湖北省自然科学基金（基金编号：2009CDA146）省部级自然科学基金

关键词：带钢表面缺陷 , independent component analysis (ICA) , sparse coding , feature extraction , defect recognition

Recognition Based on Independent Component Analysis for Surface Defects of Cold Strips

DUAN Zhi-juan1

Keywords： strip surface defects , 独立成分分析 , 稀疏编码 , 特征提取 , 缺陷识别

为了提高冷轧带钢表面缺陷识别率，提出基于独立成分分析(ICA)的缺陷图像特征提取方法。通过ICA建立缺陷图像的统计生成模型，从缺陷库中自适应地估计ICA基向量，将缺陷图像向基向量张成的空间投影，从而将图像变换到ICA域，图像在ICA域内的表示即为相应的特征向量。这种特征元素之间统计独立，是图像的稀疏编码。试验表明，本方法提取的特征优于常用的几何、纹理、不变矩特征，缺陷识别率较现有方法得到了提高。

In order to improve the recognition rate of cold rolled strip surface defects, a new method of feature extraction was investigated for defect images based on Independent Component Analysis (ICA). Base vectors were estimated adaptively from the defect library using the statistics generation model established by ICA. Defect image were transformed to ICA domain through projecting to the space spanned by base vectors. The coefficients in ICA domain were defect′s feature vector, which were statistically independent and become image′s sparse coding. Experiments show that the proposed feature is superior to geometry, texture and invariant moment features, and it produces higher recognition rate of defect images.

参考文献

上一张下一张

计量

下载量()
访问量()

作者相关

文章评分

您的评分：
1

0%
2

0%
3

0%
4

0%
5

0%

材料期刊网