A Method of Printing Press Fault Diagnosis Based on Image Texture Information

Abstract

Fault diagnosis plays an increasingly important role in the safety of printing press, and how to judge the fault quickly and accurately has aroused people’s attention. During the printing progress, substrate have contacted with many parts of printing press, which makes the printing pictures obtained many useful information, and these information can be used to diagnose printing machine faults. By analyzing these information, a new method for fault detection and diagnosis in printing press based on image texture information is presented. Firstly, the gray level co-occurrence matrix (GLCM) was used to extract the data. Secondly, the principal component analysis (PCA) method was used to process the data. Thirdly, the support vector machine (SVM) was used to analyze the data. Finally we realized the fault diagnosis with software. The availability of the diagnostic approach is verified by experiments, and we have realized the high accuracy and efficiency of fault diagnosis. It is practical in the field of printing machine fault diagnosis.