Background and the perception of lightness

Abstract

A simultaneous equisection experiment using a CRT in a dark surround was performed to investigate the relationship between a uniform background and the perception of lightness. The resulting curves for different backgrounds show both exponential properties, for black and white backgrounds, and sigmoidal characteristics, for intermediate grays. The sigmoidal properties are due to crispening and roughly intersect the diagonal or identity at the lightness of the background. The simultaneous contrast is greatest for the middle gray and decreases as the device white and black points are approached. An example equation is provided to fit the observations. This equation has as its input the L* of the stimuli and the background and has two fitting parameters. The issue of lightness scaling for backgrounds is also considered and finally extensions to this research are briefly mentioned.

The Role of Lightness Perception in Determining the Perceived Contrast of Real World Scenes

“…The results show a clear effect of the background luminance level upon the perceived contrast, and in turn, perceived image quality, of the small natural image patches. This pattern of results is broadly consistent with the established effect of the background luminance on the luminance-to-lightness function [3], however comparison with the literature on lightness perception is difficult because the exact nature of the luminance-to-lightness function appears highly sensitivity to the exact experimental conditions, as evidenced by the diversity of estimated functions [7,17,21,24,18,4,25,19,22,3]. Moreover, research on simple stimuli does not appear to generalize to the perception of real world scenes [3].…”

“…A number of studies investigate lightness perception using simple stimuli, typically composed of a few, small test patches viewed upon an extended, uniform surround. The key finding is that the background luminance level can dramatically alter the relationship between the onscreen luminance of a test patch and the perceived lightness of the patch [7,17,21,24,18,4,25,19]; In some studies the effect is dramatic, with the luminance-to-lightness function being compressive when the stimuli is viewed upon a dark background, but closer to linear, or even expansive, when viewed upon a light background [17,22,3]. Similar findings are observed when the perceived lightness of real world scenes is investigated [4], although the effect of the background luminance is less dramatic and the function is always observed to be compressive [3].…”

System gamma as a function of image- and monitor-dynamic range

“…To investigate viewing-condition dependency we investigate the preferred system gamma for images viewed on a black or white background and upon monitors; a cathode ray tube (CRT), a liquid crystal display (LCD), and an organic light-emitting diode (OLED) that have different dynamic ranges. We choose these monitors/conditions as both the background luminance condition (Moroney, 2002;Nundy & Purves, 2002;Stevens & Stevens, 1963;Whittle, 1992) and the dynamic range of the scene/monitor (Hoffman, Johnson, Kim, Vargas, & Banks, 2015;Nezamabadi, Miller, Daly, & Atkins, 2014;Radonjić et al, 2011) are known to affect the shape of the luminance-tolightness function. (Fairchild, 2007).…”

Is there a preference for linearity when viewing natural images?

“…A gamma function has several advantages (a) it has mathematical simplicity, particularly considering that system gamma can also be modeled as a gamma function and (b) a gamma function can capture the continuum from compressive to expansive non-linearities. We acknowledge that a gamma function may not correctly capture lightness perception under all conditions, [8][9][10] but note that lightness perception is highly sensitive to the exact experimental conditions and no research claims to have developed a function that can generalize to all experimental conditions; moreover, only a few studies have investigated lightness perception using natural scenes. 10,28 Finally, as we shall infer the onscreen luminance to lightness function from our dataset, rather than directly measure the function, we consider it wise to use the most general formula available.…”