It can be stated that image quality is in the eye of the beholder. After all, the human observer is the final arbiter for an imaging device whose output is intended for human consumption. However, it is difficult to distill objective data about specific image elements and attributes from image evaluation processes and procedures that rely on human observations. Quantification of image quality attributes is confounded by the inherent subjectivity of human judgment, and the fact that human perception is a complex mixture of psychology, physiology, and environment. In spite of these difficulties, the need for quantitative image quality analysis still exists. Quantification provides the basis for intersystem comparisons, evaluation of performance against specifications, and it can be a critical component in process control and failure analysis.Objective image quality evaluation systems can provide the repeatability and reliability lacking in subjective processes. A machine-vision-based system can provide detailed information about individual attributes that contribute to the overall perception of image quality. Line quality, dot quality, and color reproduction are just a few of the many elements that can be characterized in detail by such a system. In general, it is important to maintain a correlation between measured attributes and the human visual response. However, a machine-vision-based system has an additional benefit in that it can provide additional information that can be used to differentiate between multiple possible causes for a single defect.If, in addition to being objective, a measurement system is also automated, the capabilities and resulting benefits of the system increase dramatically. A well designed, automated system can support the evaluation of a large volume of prints and a large number of image quality attributes, as well as a wide variety of test targets.In this paper we will be discussing the importance of objective analysis in image quality evaluation. We will discuss several key image quality attributes such as dot and line quality, color registration, and tone reproduction, as well as address some technology-specific image quality attributes. We will present some of the metrics that can be used to quantify these attributes. We will also discuss how an automated, objective image quality measurement system can be a critical component in statistical process control and failure analysis.