A printing defect online identification method and system based on visual detection

Abstract

The present application provides a kind of printing defect online identification method and system based on visual detection, belong to printing quality control technical field.For the problem that prior art is difficult to distinguish whether printing defect is caused by dynamic defect of speed fluctuation or static defect of mechanical wear, the method extracts dot edge roughness, shoulder angle and bottom filling rate through dark field microscopic imaging, and generates printing pressure distribution map and ink transfer rate distribution map by inversion;Through encoder phase synchronization to establish drum circumferential mapping, analyze the periodic distribution mode of pressure anomaly to determine the mechanical correlation type;Further, through the cross-correlation of pressure anomaly amplitude sequence and speed fluctuation sequence and the phase difference analysis at dominant frequency, it is determined whether the defect is caused by speed fluctuation and the specific process link is traced back;Finally, combined with the degradation trend of dot microstructure parameters, the replacement of printing plate warning, drum dynamic balance adjustment or ink supply pressure correction grading processing instruction is generated.

Images