Digital video quality metric based on human vision

Citation
Ab. Watson et al., Digital video quality metric based on human vision, J ELECTR IM, 10(1), 2001, pp. 20-29
Citations number
24
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Optics & Acoustics
Journal title
JOURNAL OF ELECTRONIC IMAGING
ISSN journal
1017-9909 → ACNP
Volume
10
Issue
1
Year of publication
2001
Pages
20 - 29
Database
ISI
SICI code
1017-9909(200101)10:1<20:DVQMBO>2.0.ZU;2-L
Abstract
The growth of digital video has given rise to a need for computational meth ods for evaluating the visual quality of digital video. We have developed a new digital video quality metric, which we call DVQ (digital video quality ) [A. B. Watson, in Human Vision, Visual Processing, and Digital Display VI II, Proc. SPIE 3299, 139-147 (1993)]. Here, we provide a brief description of the metric, and give a preliminary report on its performance. DVQ accept s a pair of digital video sequences, and computes a measure of the magnitud e of the visible difference between them. The metric is based on the discre te cosine transform. It incorporates aspects of early visual processing, in cluding light adaptation, luminance, and chromatic channels; spatial and te mporal filtering; spatial frequency channels; contrast masking; and probabi lity summation. It also includes primitive dynamics of light adaptation and contrast masking. We have applied the metric to digital video sequences co rrupted by various typical compression artifacts, and compared the results to quality ratings made by human observers. (C) 2001 SPIE and IS&T.