Tone Reproduction

in photography, the transmission to a photographic image of the differences in brightness (B) of details in the subject photographed. The differences are rendered as differences in optical density (D) in the image.

With ideal photographic tone reproduction in the final positive image, whether a print or diapositive, the differences in D for any two details should be such that the ratio of brightnesses in the subject and the image are identical under identical conditions of examination. Such tone reproduction is effected only if the entire sequence of conversions is linear and if the range in which the conversions take place is unlimited. The entire sequence includes the conversions of the set of brightnesses to the set of illuminations on the negative emulsion and, subsequently, to negative blackening, illumination on the positive emulsion, positive blackening, and so on, to the set of blackenings on duplicates or the set of illuminations on a viewing screen. In practice, however, all the stages of the photographic process are more or less nonlinear and are limited in range. For example, the characteristic curve of the photosensitive emulsion is always nonlinear, and its exposure latitude is limited and insignificant, even compared to the complete range of transmitted exposures.

Tone reproduction is complicated even more if the image is evaluated visually, rather than by means of objectively measured characteristics. In this case, the number of distorting factors is increased to include the characteristics of the eye as a receiver, which may evaluate the same set of values of B differently, depending on the size of the details, the various illumination levels of the image and the subject, and uneven darkening in the room used for examination. As a result, the basic problems in the theory of tone reproduction are no longer seen as the determination of conditions for ideal tone reproduction. Rather, by necessity, they have gradually been reformulated as the determination of conditions under which tone reproduction is still satisfactory for the possible wider range of values B as the analysis of how to reproduce a given subject photographically with the smallest degree of distortion of the differences in B, and as the determination of quantitative estimates of such distortion.