Share Email Print

Proceedings Paper

Perceptual metrics and visualization tools for evaluation of page uniformity
Author(s): Minh Q. Nguyen; Renee Jessome; Steve Astling; Eric Maggard; Terry Nelson; Mark Shaw; Jan P. Allebach
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Uniformity is one of the issues of most critical concern for laser electrophotographic (EP) printers. Typically, full coverage constant-tint test pages are printed to assess uniformity. Exemplary nonuniformity defects include mottle, grain, pinholes, and “finger prints". It is a real challenge to make an overall Print Quality (PQ) assessment due to the large coverage of a letter-size, constant-tint printed test page and the variety of possible nonuniformity defects. In this paper, we propose a novel method that uses a block-based technique to analyze the page both visually and metrically. We use a grid of 150 pixels × 150 pixels ( ¼ inch × ¼ inch at 600-dpi resolution) square blocks throughout the scanned page. For each block, we examine two aspects: behavior of its pixels within the block (metrics of graininess) and behavior of the blocks within the printed page (metrics of nonuniformity). Both ΔE (CIE 1976) and the L* lightness channel are employed. For an input scanned page, we create eight visual outputs, each displaying a different aspect of nonuniformity. To apply machine learning, we train scanned pages of different 100% solid colors separately with the support vector machine (SVM) algorithm. We use two metrics as features for the SVM: average dispersion of page lightness and standard deviation in dispersion of page lightness. Our results show that we can predict, with 83% to 90% accuracy, the assignment by a print quality expert of one of two grades of uniformity in the print.

Paper Details

Date Published: 3 February 2014
PDF: 13 pages
Proc. SPIE 9016, Image Quality and System Performance XI, 901608 (3 February 2014); doi: 10.1117/12.2038752
Show Author Affiliations
Minh Q. Nguyen, Purdue Univ. (United States)
Renee Jessome, Hewlett-Packard Co. (United States)
Steve Astling, Hewlett-Packard Co. (United States)
Eric Maggard, Hewlett-Packard Co. (United States)
Terry Nelson, Hewlett-Packard Co. (United States)
Mark Shaw, Hewlett-Packard Co. (United States)
Jan P. Allebach, Purdue Univ. (United States)

Published in SPIE Proceedings Vol. 9016:
Image Quality and System Performance XI
Sophie Triantaphillidou; Mohamed-Chaker Larabi, Editor(s)

© SPIE. Terms of Use
Back to Top