一种基于机器视觉的纸病识别方法  

A Paper Defects Detection Method Based on Machine Vision

作　　者：张学兰[1,2] 李军[1,2] 孟范孔[3] 

机构地区：[1]华南理工大学制浆造纸工程国家重点实验室,广东广州510640 [2]华南理工大学广东省造纸技术与装备公共实验室,广东广州510640 [3]华南理工大学机械与汽车工程学院,广东广州510640

出　　处：《中国造纸学报》2013年第1期48-52,共5页Transactions of China Pulp and Paper

基　　金：国家重点基础研究发展规划(973计划;2010CB732205);国家科技支撑计划项目(2007BAF25B00)

摘　　要：提出了一种基于图像灰度变换和BP神经网络的纸病识别算法。该算法是利用动态双阈值法和图像合成法对图像进行预处理,提取出纸病的特征值,再利用BP神经网络对纸病进行分类。结果表明,BP神经网络分类器可以很好地识别出纸病图像中的孔洞、脏点和褶皱,平均识别率达93.8%。A new paper defects recognition algorithm based on image gray transformation and BP neural network was put forward.The paper image was preprocessed,then the paper defects characteristic value was extracted,finally,the BP neural network was used to classify paper defects.Experimental results showed that this algorithm could successfully recognize a paper image that contains holes,spots and folds.The precision of the performance of the system reached 93.8%.

关 键 词：纸病图像 图像灰度 特征值 BP神经网络 

分 类 号：TS736.2[轻工技术与工程—制浆造纸工程]