Lightness, Brightness and Transparency

by Alan L. Gilchrist (Editor)

This volume deals with the visual perception of lightness, brightness, and transparency of surfaces, both under minimal laboratory conditions and in complex images typical of everyday life. Each chapter analyzes the challenging problem of how a pattern of light intensities on the retina is transformed into the visual experience of varying shades of grey, transparent surfaces, and light and shadow. One important theme which unifies the group of contributions is the recognition that the perception of surface lightness is rooted fundamentally in the encoding of relative intensities of light within the retinal image, not intensities per se. A second important unifying theme is an appreciation of the multiple dimensions of the visual experience of lightness, brightness, and transparency -- people do not perceive the lightness of surfaces by discarding information concerning the light illuminating those surfaces; rather, they perceive a pattern of illumination projected onto a pattern of surface greys.

The long-fascinating problems of surface lightness and color perception have become very active topics recently as a resurging interest within the visual perception community has coincided with an increasing appreciation of the centrality of these problems by the emerging machine vision community. The best of recent psychophysical work on lightness perception, as presented in this volume, will be of great interest to both of these communities. This book also marks a synthesis of old and new. A traditional, strongly Gestalt, approach that had fallen into neglect is updated in the light of new quantitative systematic methods and important later discoveries, such as the disappearance of stabilized retinal images. The book draws on such diverse approaches as Gestalt and ecological psychology, threshold psychophysics, and computational vision, advancing our understanding of the interrelations among surface color, illumination, perceived depth, shading, and transparency.

330 pages, Hardcover

First published September 1, 1994

About the author

Alan Gilchrist is a Professor in the Faculty of Psychology at Rutgers University in the USA.

"I study visual perception, especially the perception of surface color, and especially the black-white dimension. Vision is known to be based on the image projected onto the retina, but the problem of how to assign black, white and gray values to surfaces represented in that image remains unsolved, in human vision as in computer vision. Because of variations in many factors such as the background of a surface and the lighting conditions, the perception of any one specific surface color can be associated with many patterns of local stimulation at the retina. The goal of the work is to describe the software (not the hardware, or wetware) used by the visual system to decode the retinal image. The primary method is psychophysics. Naive observers are exposed to displays specially constructed so that competing theories make opposing predictions of what observers will see. The observer reports, typically involving matches made using a color chart, are then used to evaluate theories. In my lab we have approached this problem in two ways. In earlier work, an inverse-optics approach was taken in which we attempted to determine the computations necessary to recover objective properties like surface color. More recent work has focused on the pattern of errors shown by human observers when judging surface colors. These errors are systematic, not random, and the work is based on the assumption that the pattern of errors is the signature of the software used to decode the retinal image."