Classifying digital prints according to their production process using image analysis and artificial neural networks
Jack Tchan · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2000
A human expert observer can be employed to identify the production source of a print. The observer achieves this task by visual inspection of the print using a microscope. However, there are cases when the expert observer fails to identify correctly the production source. It is for this reason that the development of a method which can identify the production source is under consideration. This paper discusses the initial stages of the project which focuses on the development of a system that can classify prints from three different digital printing process. The system comprised an image analyzer that supplied image data from the print samples for initial analysis using a data pre- processing program and artificial neural networks which then used the pre-processed data to produce the classification models. The three different digital printing processes employed in this investigation were laser printing, optical photocopying and inkjet printing. Print samples were obtained from a range of laser printers,,optical photocopiers and inkjet printers. The prints used in the investigation were of a monochrome image of a square. The results show that the system is capable of classifying prints accurately for the range of printing machines and the image used in the trials.