Objective Picture Quality Measurements: Expectations Today and Tomorrow

In addition to traditional indirect signal quality measurements, today's objective picture quality measurements require analysis of picture changes caused by nonlinear and time-variant digital processing systems such as compression codecs. Although the objective measurement paradigm, based for the first time on estimating directly perceived changes in quality, is new, we can expect much the same confidence in system operation as today. For broad utility, these new objective measurements must be meaningful to the application, and generally have correspondence to subjective evaluation of the changes in picture quality caused by video processing. Objective measurements can also be expected to have advantages in providing much more rapid and consistent results over a wide range of picture degradations. As a guide to understanding the various attributes that are desirable for robust objective measurement methods, we discuss appropriate methods for correlating objective measurements with subjective results. The need for standardized test scenes is explained as well as considerations for in-service monitoring applications.

Published

1998-10

Content type

Original Research

DOI

10.5594/M00300

ISBN

978-1-61482-929-4