Luminance-based specular gloss characterization

Abstract

Gloss is a feature of visual appearance that arises from the directionally selective reflection of light incident on a surface. Especially when a distinct reflected image is perceptible, the luminance distribution of the illumination scene above the sample can strongly influence the gloss perception. For this reason, industrial glossmeters do not provide a satisfactory gloss estimation of high-gloss surfaces. In this study, the influence of the conditions of illumination on specular gloss perception was examined through a magnitude estimation experiment in which 10 observers took part. A light booth with two light sources was utilized: the mirror image of only one source being visible in reflection by the observer. The luminance of both the reflected image and the adjacent sample surface could be independently varied by separate adjustment of the intensity of the two light sources. A psychophysical scaling function was derived, relating the visual gloss estimations to the measured luminance of both the reflected image and the off-specular sample background. The generalization error of the model was estimated through a validation experiment performed by 10 other observers. In result, a metric including both surface and illumination properties is provided. Based on this metric, improved gloss evaluation methods and instruments could be developed.

Why do rough surfaces appear glossy?

“…Researchers have suggested several cues observers may use to infer surface gloss and most previous work has concentrated on specular highlights and DOI gloss [2,15,[23][24][25][26][27][28]. In our experiments, the microscale parameters were specifically chosen so that DOI gloss was not obvious, and thus for our surfaces it is the specular highlights that are likely to provide the most important cues.…”

A unified account of gloss and lightness perception in terms of gamut relativity

“…According to this equation-which, when evaluated in terms of luminance values, is closely related to some recently proposed gloss metrics[20]-surface gloss ranges from 0 when s s ν to 1 when js − s ν j js w − s b j. As shown in Figs.…”

“…According to theoretical approaches based on the classical computational view of vision [4][5][6], the visual system computes lightness and gloss at each image location as a function of spatial image gradients correlated to diffuse and specular reflectance components, respectively [7][8][9][10][11]. The modularity inherent in this notion has spawned a surfeit of perceptually and computationally formulated theories of gloss and lightness perception [9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]-each with a limited explanatory rangerather than a unified, predictive theory.…”

Overall gloss evaluation in the presence of multiple cues to surface glossiness

“…In order to investigate and to formulate measurement alternatives that quantify perceived gloss better than the specular glossmeter does, Leloup et al picked up the luminance-based approach that vision scientists put forward [27]. Indeed, while the specular and diffuse reflectance are attributes of a material, the luminance distribution perceived when looking at a surface is determined by the surface characteristics and by the characteristics of the illuminating field surrounding the sample.…”