Ghosting is a well-known Print Quality (PQ) defect which may appear as a single or repetitive artifact presenting vestigial objects at a certain interval. The overall print quality will be limited if ghosting is present. Therefore, an algorithm that can accurately provide the information of both the ghosting source and its severity is greatly needed. In this paper, we propose an algorithm to detect and evaluate ghosting by first applying template matching in the CIE L*a*b* color space, and then calculating the color difference. The template matching step in the L* channel will indicate the position and the type (light or dark) of the ghosting. We then calculate the color difference among L*, a*, and b* channels to get the Delta E for the purpose of evaluation. Our algorithm can automatically detect, quantify, and label the severity of ghosting according to a final metric. Base on 82 samples in total, the accuracy of our algorithm is 92% compared with expert visual evaluation. Our algorithm is also suitable to be used as a quality control tool to set limits in production processing.
Xiaochen Jing, Steve Astling, Renee Jessome, Eric Maggard, Terry Nelson, Mark Shaw, Jan P. Allebach, "Electrophotographic Ghosting Detection and Evaluation" in Proc. IS&T Int'l Conf. on Digital Printing Technologies and Digital Fabrication (NIP31), 2015, pp 169 - 172, https://doi.org/10.2352/ISSN.2169-4451.2015.31.1.art00037_1