Toner usage prediction
Abstract
In recent years, laser electro-photographic (EP) printers are commonly used in both industry and households. From an economic perspective, it is of great significance to accurately estimate toner usage in order to get a full utilization of each cartridge. A revised two-stage strategy is developed based on the Black Box Model (BBM) which demonstrates that the toner deposition in the area occupied by each printer-addressable pixel is strongly influenced by the values of the neighboring pixels. For the first stage, the pixel-by-pixel absorptance is predicted from the digital page that is sent to the laser engine, and trained from a set of printed and scanned pages. Then, during stage two, the overall toner usage for input digital page can be estimated by using a weighted sum method that adds up the weight of every individual pixel of the whole page. The weights of pixels are trained by using the measured toner usage of each page in a set of training pages, and the histogram of the pixel values for the predicted scanned pages in the training set. Compared with the pixel counting method and our previous two-stage predictor, the updated approach is not only more robust and accurate, but also more suitable for hardware design.
Degree
M.S.E.C.E.
Advisors
Allebach, Purdue University.
Subject Area
Electrical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.