Abstract
The evolution of photographic, optical, electronic or hybrid imaging systems has required the application of a number of physical measures of image quality. These measures are needed, to assess any improvements made, to provide aim points for improvement, to provide the user of the system with the means of comparing systems and for obtaining optimum results for a particular imaging task. However, image quality has no single unique measure or definition. This is not surprising in view of the complex nature of the visual process even when concepts of quality are restricted to the simplest case of two-dimensional, stationary, monochrome images, such as photographs or images on a monitor, and become even more complex when colour is included.
This paper takes an overview of physical attributes of image quality which include: contrast, resolution, sharpness, noise and colour, and gives details of how these quantities can be correlated with psychometric measures of image quality, determined by panels of end-users. The objectives being to provide a general system for the evaluation of image quality such that imaging systems can be compared with one another and that changes brought about in a single system, by image processing for example, can be quantified from the end-users view-point.