摘要
打印文件检验在公共安全领域有着重要的作用。研究了基于打印墨粉图像分析的计算机打印文件检验,以快速鉴别打印文件的源打印机。鉴别过程包括打印墨粉堆积图像采集,图像预处理,特征提取和匹配。用自行设计的装置采集打印文件图像,经过打印墨粉图像归一化构造纹理图像,利用灰度共生矩阵和频谱方法对不同打印文件中的打印墨粉堆积纹理进行匹配,来进行打印文件的源打印机认定。在有45台激光打印机的数据库中测试本文方法,效果良好,证明了本文方法的有效性。
Printed document examination plays an important role in public security.A study on print toner image analysis based printed document examination is presented in purpose of identifying the printer which created a suspect printed document.The procedure includes image normalization,feature extraction and feature matching.Uniform texture images are constructed in image normalization; the Fourier transform technique and Gray level co-occurrence matrix technique are adopted for feature extraction; feature selection and feature fusion are implemented next,finally the nearest-neighbor classifier based on Euclidean distance measure is adopted for identification.Experiments are carried out in a database including 45 LaserJet printers and very promising results are achieved.The proposed method provides a text-independent way for the research of forensic printed document examination system.
出处
《电子测量技术》
2014年第2期70-74,90,共6页
Electronic Measurement Technology
关键词
打印文件检验
纹理分析
文本独立
傅里叶变换
printed document examination
texture analysis
text-independent
Fourier transform
