Share Email Print

Proceedings Paper

Comparison of vision-based algorithms for hiding defective sub-pixels
Author(s): Joe Stellbrink
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

The potential use of vision-based algorithms for hiding defective display pixels is quite appealing. Two prior approaches utilized either the point spread function (PSF) or contrast sensitivity functions to represent effects of the human visual system. A third approach proposed in this paper includes a simple model of human visual masking characteristics to improve theoretical defect hiding effectiveness. A visual experiment indicated all three methods provided significant improvement over uncompensated sub-pixel defects across all color patches and images tested. The masking-based method and an empirically optimized PSF method were more effective due to the masking-type patterns generated. Hiding effectiveness was linearly related to the inverse of the lightness error generated by a defect. For moderate lightness errors, both the PSF and masking-based methods completely hid the sub-pixel defects, with decreasing effectiveness for larger lightness errors. Similar results were found for images and corresponding color patches, though some dependency on the image content was observed for two of the five images. With the addition of a simple visual masking effects model, the iCAM Image Difference Model was found to predict the general performance trends of the three methods with reasonable accuracy.

Paper Details

Date Published: 29 January 2007
PDF: 12 pages
Proc. SPIE 6494, Image Quality and System Performance IV, 64940Q (29 January 2007); doi: 10.1117/12.704336
Show Author Affiliations
Joe Stellbrink, Hewlett-Packard Co. (United States)

Published in SPIE Proceedings Vol. 6494:
Image Quality and System Performance IV
Luke C. Cui; Yoichi Miyake, Editor(s)

© SPIE. Terms of Use
Back to Top