Densitometry provides the method and instrumentation for determining the optical density of objects. There are two types of optical density measurements: transmission and reflection. Transmission densitometry measures the density of transmissive samples such as the 35-mm film and overhead transparency. Reflection densitometry measures the density of reflected samples such as photographic and offset prints. Densitometry is widely used in the printing industry because most forms of color encoding are based directly or indirectly on density readings; most input and output devices are calibrated by using density measurements, and most reflection and transmission scanners are essentially densitometers or can be converted into one.
Optical densities of objects correspond quite closely to human visual sensibility. Therefore, density is a powerful measure of the object's color quality because, within a relatively wide range, the density follows the proportionality and additivity of the colorant absorptivity. In theory, one can use the additivity law to predict the density of a mixture from its components. However, there are problems in achieving this goal because of the dependencies on instrumentation, imaging device, halftone technique, and media. Nevertheless, densitometry is widely used in the printing industry for print quality control and other applications. This is because of its simplicity, convenience, ease of use, and an approximately linear response to human visual sensibility. In the foreseeable future, the use of densitometry is not likely to decrease. Therefore, we present the fundamentals of densitometry, its problems, and its applications.
Perhaps the most important application is the color reproduction that uses the density masking method based on density additivity and proportionality. We present the derivation of the density-masking equation and examine the validity of the assumptions. We then extend the density-masking equation to the device-masking equation. Modifications and adjustments of tone characteristics are suggested to fit the assumptions better. We propose several methods for deriving coefficients of the device-masking equation by using the measured densities of primary color patches. The advantages and disadvantages of these methods are discussed. Digital implementations of the density and device-masking equations are given in Appendix 9.
Online access to SPIE eBooks is limited to subscribing institutions.