Luminance Gradients Research Articles

Uniformity perception for stimulus with luminance gradient

Uniformity perception for stimulus with luminance gradient

Luminance gradient configuration determines perceived lightness in a simple geometric illusion.

Accurate perception of surface reflectance poses a significant computational problem for the visual system. The amount of light reflected by a surface is affected by a combination of factors including the surface’s reflectance properties and illumination conditions. The latter are not limited by the strength of the illuminant but also include the relative placement of the light illuminating the surface, the orientation of the surface and its 3d shape, all of which result in a pattern of luminance gradients across the surface. In this study we explore how luminance gradients contribute to lightness perception. We introduce a novel, simple lightness illusion. It consists of six separate checks, organized in rows of two. Each check has a negative luminance gradient across it. The top and the bottom rows are the same: with the darker check on the left, and the lighter check on the right. Two checks in the middle row are identical; however, the check on the right appears darker than the check on the left. As there are no shared borders between the checks, simultaneous contrast cannot explain the effect. However, there are multiple possible explanations including spatial filtering (Blakeslee and McCourt, 2004) or some higher-order mechanism such as perceptual grouping or amodal completion. Here, we explore these possibilities by manipulating the luminance configurations and the gradient slopes of the checks.

A temporal window for estimating surface brightness in the Craik-O'Brien-Cornsweet effect.

The central edge of an opposing pair of luminance gradients (COC edge) makes adjoining regions with identical luminance appear to be different. This brightness illusion, called the Craik-O'Brien-Cornsweet effect (COCe), can be explained by low-level spatial filtering mechanisms (Dakin and Bex, 2003). Also, the COCe is greatly reduced when the stimulus lacks a frame element surrounding the COC edge (Purves et al., 1999). This indicates that the COCe can be modulated by extra contextual cues that are related to ideas about lighting priors. In this study, we examined whether processing for contextual modulation could be independent of the main COCe processing mediated by the filtering mechanism. We displayed the COC edge and frame element at physically different times. Then, while varying the onset asynchrony between them and changing the luminance contrast of the frame element, we measured the size of the COCe. We found that the COCe was observed in the temporal range of around 600–800 ms centered at the 0 ms (from around −400 to 400 ms in stimulus onset asynchrony), which was much larger than the range of typical visual persistency. More importantly, this temporal range did not change significantly regardless of differences in the luminance contrast of the frame element (5–100%), in the durations of COC edge and/or the frame element (50 or 200 ms), in the display condition (interocular or binocular), and in the type of lines constituting the frame element (solid or illusory lines). Results suggest that the visual system can bind the COC edge and frame element with a temporal window of ~1 s to estimate surface brightness. Information from the basic filtering mechanism and information of contextual cue are separately processed and are linked afterwards.

Luminance gradient configuration determines perceived lightness in a simple geometric illusion

Accurate perception of surface reflectance poses a significant computational problem for the visual system. The amount of light reflected by a surface is affected by a combination of factors including the surface’s reflectance properties and illumination conditions. The latter are not limited by the strength of the illuminant but also include the relative placement of the light illuminating the surface, the orientation of the surface and its 3d shape, all of which result in a pattern of luminance gradients across the surface. In this study we explore how luminance gradients contribute to lightness perception. We introduce a novel, simple lightness illusion. It consists of six separate checks, organized in rows of two. Each check has a negative luminance gradient across it. The top and the bottom rows are the same: with the darker check on the left, and the lighter check on the right. Two checks in the middle row are identical; however, the check on the right appears darker than the check on the left. As there are no shared borders between the checks, simultaneous contrast cannot explain the effect. However, there are multiple possible explanations including spatial filtering (Blakeslee and McCourt, 2004) or some higher-order mechanism such as perceptual grouping or amodal completion. Here, we explore these possibilities by manipulating the luminance configurations and the gradient slopes of the checks.

Effects of delay and distractors in temporal search on clips of fire

Fire has been a ubiquitous natural visual stimulus throughout evolution. The swiftly-changing colour and luminance gradients, and the variety of shapes poses a challenge for dynamic form encoding. Here, we investigated human capacity for dynamic form encoding and recognition, testing the effect of delay and padding on temporal search for dynamic targets within fire clips. In Experiment 1 we used a delayed-match-to-sample task with clips drawn from video of a hearth fire. Subjects viewed a 1-second sample clip followed by a longer test clip, then judged whether the sample was in the test (yes/no). In Experiment 1, we varied the delay between sample and test (1,5,10,15 seconds), and clip orientation (upright, inverted). Performance decreased linearly with time from 67% correct (delay=1 second) to 58% correct (delay=15 seconds) (p<0.05, paired-samples t-test). Performance was not significantly different for inverted clips, showing a drop from 67% to 66% (p=0.3, paired-samples t-test). In Experiment 2 we adjusted the temporal position of the 1-second sample in the test clip (test-sample delay 1 second). The sample was preceded by a clip of length x and followed by a clip of length y. We varied x and y independently across {0.25, 0.75, 1.25, 2.25} seconds. The test clip thus lasted x+s+y seconds, whether the sample was present or not. Increasing the length of the post-stimulus padding from 0.25 s to 2.25 s caused a large drop in mean accuracy (from 75% to 52%, averaged across all pre-padding lengths; p=0.0073 by paired-sample t-test). We conclude that detection of target clips of fire degrades linearly with sample-test delay, but is not substantially degraded by inversion (Experiment 1). Furthermore, distractors cause error even when presented after the sample (Experiment 2), showing that dynamic samples of fire are not easily individuated.Meeting abstract presented at VSS 2014

The illusory brightening MAE separates low-level motion models.

The brightening MAE provides a challenge to low-level motion models. Introduced by Anstis (1967, Science 155 710-712), uniform fields appear to brighten after adaptation to repeated darkening. When the uniform test field is replaced by a spatial gradient, observers report illusory motion. We first measured the apparent speed of the aftereffect. Observers adapted to a circular pattern of linearly lightening or darkening radial segments, which were later replaced by static luminance gradients. When brightening or darkening, the adjacent segments of the dynamic pattern were set to grey so that any local motion energy at the boundaries was balanced over the adaptation period. At test, the sign of the spatial gradient was chosen appropriately for each segment to deliver uniform rotational motion. We found that the perceived speed of the rotation was precisely inversely proportional to the spatial gradient, as predicted by a spatiotemporal gradient motion computation. Motion energy models employ spatially band-pass front-end filters and therefore will not respond to uniform fields (the even Gabor is an exception). However, if the temporal brightening and spatial luminance gradient were combined before motion analysis, motion at test might be predicted by alternative models. To separate the models we sculpted the profile, introducing non-zero higher-order spatial derivatives. We now observed an apparent slowing of the aftereffect, which increased in turn with the accumulation of higher orders of spatial derivative for both convex and concave luminance ramps. Crucially, computer simulation shows the energy model and the multichannel gradient model predict increased speed from a simple addition of temporal brightening and the test spatial pattern. The data however are consistent with an isolated spatially low-pass first temporal derivative component delivering brightening, divided by multiple spatial derivatives introduced in the test pattern, a previously proposed multichannel gradient architecture, resulting in slowing of perceived speed.Meeting abstract presented at VSS 2014

Case studies of a camera-aided imaging method for evaluation of interior luminous environments

This paper introduces a camera-aided imaging method to help lighting practitioners, especially students and beginners, to evaluate the quality of interior luminous environments. This imaging method adopts the high dynamic range photography to acquire millions of luminance values across an interior luminous environment, stores the data on a high dynamic range (HDR) image, and then uses a MatLab code ( http://people.ku.edu/∼h717c996/research.html ) to plot two-dimensional luminance and luminance gradient maps of the luminous environment for further evaluation. Four case studies were conducted to validate this imaging method. The HDR image of each test scenario and its luminance and luminance gradient maps were analysed to evaluate the lighting quality in light of four visual concepts, including hierarchy of focal accents, glare and reflectance, entry of daylight, and lighting the walls. Information extracted from those images could be used to improve problematic lighting conditions. The method is useful for vision-based lighting evaluation practice.

Motion influences the perception of background lightness.

Uniform backgrounds appear lighter or darker when elements containing luminance gradients move across them, a phenomenon first presented by Ko Nakamura at the 2010 Illusion Contest in Japan. We measured the apparent lightness of the background with a configuration where the grey background was overlaid with moving square patches of vertically oriented luminance gradient. For black-to-grey gradients, the background appeared lighter when the black edges were leading than when they were trailing. For white-to-grey gradients, the background appeared darker when the white edges were leading than when they were trailing. For white-to-black gradients, the background appeared darker with a white edge leading and lighter with a dark edge leading, but the effects were weaker. These results demonstrate that lightness contrast can be modulated by the direction of motion of the inducing patterns. The smooth gradient is essential, because the effect disappeared when the black-to-white gradient was replaced with the binary black and white pattern. We speculate that asymmetry in the processing of a temporal gradient with increasing and decreasing contrast, as proposed to explain the “Rotating Snakes” illusion (Murakami, Kitaoka, & Ashida, 2006, Vision Research, 46, 2421–2431), might be the basis for this effect.

Luminance gradient for evaluating lighting

This paper introduces luminance gradient as a metric for evaluating common non-uniform luminous environments. Luminance gradient is defined as the largest change rate and the polarity of spatial luminance variation on a large surface or across the entire field of view in high resolution. The metric, a visualization method, and a freely available MatLab code for plotting luminance gradient maps have been developed. The retinal resolution of pixels for mapping luminance gradient has also been identified. A camera-aided measurement method of luminance gradient is proposed, which is demonstrated in two case studies of the evaluation of different types of luminous environments. Moreover, this study explored whether luminance gradient could also be obtained from computer simulations using lighting software programs AGI32 and HeliosPro.

Perceived depth from luminance gradient and disparity

Perceived depth from luminance gradient and disparity

Enhancement of Backlight-Scaled Images

Switching the liquid crystal display (LCD) backlight of a portable multimedia device to a low power level saves energy but results in poor image quality especially for the low-luminance image areas. In this paper, we propose an image enhancement algorithm that overcomes such effects of dim LCD backlight by taking the human visual property into consideration. It boosts the luminance of image areas below the perceptual threshold while preserving the contrast of the other image areas. We apply the just noticeable difference theory and decompose an image into an HVS response layer and a background luminance layer. The boosting and compression processes, which enhance the visibility of the low-luminance image areas, are carried out in the background luminance layer to avoid luminance gradient reversal and over-compensation. The contrast of the processed image is further enhanced by exploiting the Craik-O'Brein-Cornsweet visual illusion. Experimental results are provided to show the performance of the proposed algorithm.

Do sandhoppers use the skylight polarization as a compass cue?

Investigations carried out in the 1950s on the use of polarized light by sandhoppers as a compass reference did not reach convincing results. Therefore, polarized light compass orientation in the sandhopper Talitrus saltator seems to be a traditional attribute more than a solid demonstration. Our tests carried out under artificial and natural conditions of illumination using different polarizing and neutral density (grey) filters demonstrated that T. saltator perceives polarized light; however, it does not use the skylight polarized light as a cue for menotactic orientation. However, since the sky spectral and luminance gradients were not completely eliminated in our tests, a possible influence of these two celestial factors of orientation could still be present. Therefore, we conclude that this species, in this experimental paradigm at least, does not use a polarized light compass mechanism, despite a very good sun compass system of orientation.

P.24: Relation between Recognition of Illumination and Depth Perception

AbstractWe investigated quantitative relation between shading on the surface and the perceptual evaluation of depth using primary and secondary cylindrical objects using 2D and 3D displays. Results showed that luminance gradient of the secondary object that gives most natural protrusive perception as a cylinder is the same as that of the primary object. Interestingly, the cases where luminance of brighter part of the secondary stimulus is relatively low, they are less accepted when it is supposed to be in front of the primary object.

Frames as visual links between paintings and the museum environment: an analysis of statistical image properties

Frames provide a visual link between artworks and their surround. We asked how image properties change as an observer zooms out from viewing a painting alone, to viewing the painting with its frame and, finally, the framed painting in its museum environment (museum scene). To address this question, we determined three higher-order image properties that are based on histograms of oriented luminance gradients. First, complexity was measured as the sum of the strengths of all gradients in the image. Second, we determined the self-similarity of histograms of the orientated gradients at different levels of spatial analysis. Third, we analyzed how much gradient strength varied across orientations (anisotropy). Results were obtained for three art museums that exhibited paintings from three major periods of Western art. In all three museums, the mean complexity of the frames was higher than that of the paintings or the museum scenes. Frames thus provide a barrier of complexity between the paintings and their exterior. By contrast, self-similarity and anisotropy values of images of framed paintings were intermediate between the images of the paintings and the museum scenes, i.e., the frames provided a transition between the paintings and their surround. We also observed differences between the three museums that may reflect modified frame usage in different art periods. For example, frames in the museum for 20th century art tended to be smaller and less complex than in the two other two museums that exhibit paintings from earlier art periods (13th–18th century and 19th century, respectively). Finally, we found that the three properties did not depend on the type of reproduction of the paintings (photographs in museums, scans from books or images from the Google Art Project). To the best of our knowledge, this study is the first to investigate the relation between frames and paintings by measuring physically defined, higher-order image properties.

Mach bands and multiscale models of spatial vision: The role of first, second, and third derivative operators in encoding bars and edges

Ernst Mach observed that light or dark bands could be seen at abrupt changes of luminance gradient in the absence of peaks or troughs in luminance. Many models of feature detection share the idea that bars, lines, and Mach bands are found at peaks and troughs in the output of even-symmetric spatial filters. Our experiments assessed the appearance of Mach bands (position and width) and the probability of seeing them on a novel set of generalized Gaussian edges. Mach band probability was mainly determined by the shape of the luminance profile and increased with the sharpness of its corners, controlled by a single parameter (n). Doubling or halving the size of the images had no significant effect. Variations in contrast (20%-80%) and duration (50-300 ms) had relatively minor effects. These results rule out the idea that Mach bands depend simply on the amplitude of the second derivative, but a multiscale model, based on Gaussian-smoothed first- and second-derivative filtering, can account accurately for the probability and perceived spatial layout of the bands. A key idea is that Mach band visibility depends on the ratio of second- to first-derivative responses at peaks in the second-derivative scale-space map. This ratio is approximately scale-invariant and increases with the sharpness of the corners of the luminance ramp, as observed. The edges of Mach bands pose a surprisingly difficult challenge for models of edge detection, but a nonlinear third-derivative operation is shown to predict the locations of Mach band edges strikingly well. Mach bands thus shed new light on the role of multiscale filtering systems in feature coding.

A Quantitative Test of the Predicted Relationship between Countershading and Lighting Environment

Countershading, a vertical luminance gradient from a dark back to a light belly, is perhaps the most common coloration phenotype in the animal kingdom. Why? We investigated whether countershading functions as self-shadow concealment (SSC) in ruminants. We calculated "optimal" countershading for SSC by measuring illumination falling onto a model ruminant as a function of time of day and lighting environment. Calibrated images of 114 species of ruminant were compared to the countershading model, and phylogenetic analyses were used to find the best predictors of coats' countershading characteristics. In many species, countershading was close to the model's prediction of "optimal" countershading for SSC. Stronger countershading was associated with increased use of open lighting environments, living closer to the equator, and small body size. Abrupt transitions from dark to light tones were more common in open lighting environments but unassociated with group size or antipredator behavior. Though the SSC hypothesis prediction for stronger countershading in diurnal species was not supported and noncountershaded or reverse-countershaded species were unexpectedly common, this basic pattern of associations is explained only by the SSC hypothesis. Despite extreme variation in lighting conditions, many terrestrial animals still find protection from predation by compensating for their own shadows.

What is binocular fusion? Multiplicative combination of luminance gradients via the geometric mean

What is binocular fusion? Multiplicative combination of luminance gradients via the geometric mean

Microsaccades and blinks trigger illusory rotation in the "Rotating Snakes" illusion

Certain repetitive arrangements of luminance gradients elicit the perception of strong illusory motion. Among them, the Snakes Illusion has generated a large amount of interest in the visual neurosciences, as well as in the public. Prior evidence indicates that the Rotating Snakes illusion depends critically on eye movements, yet the specific eye movement types involved and their associated neural mechanisms remain controversial. According to recent reports, slow ocular drift--a nonsaccadic type of fixational eye movement--drives the illusion, whereas microsaccades produced during attempted fixation fail to do so. Here, we asked human subjects to indicate the presence or absence of rotation during the observation of the illusion while we simultaneously recorded their eye movements with high precision. We found a strong quantitative link between microsaccade and blink production and illusory rotation. These results suggest that transient oculomotor events such as microsaccades, saccades, and blinks, rather than continuous drift, act to trigger the illusory motion in the Rotating Snakes illusion.

What is binocular fusion? Multiplicative combination of luminance gradients via the geometric mean

What is binocular fusion? Multiplicative combination of luminance gradients via the geometric mean

The skylight gradient of luminance helps sandhoppers in sun and moon identification

To return to the ecologically optimal zone of the beach, the sandhopper Talitrus saltator (Montagu) maintains a constant sea-land direction based on the sun and moon compasses. In this study, we investigated the role of the skylight gradient of luminance in sun and moon identification under natural and artificial conditions of illumination. Clock-shifted (inverted) sandhoppers tested under the sun (during their subjective night) and under the full moon (during their subjective day) exhibit orientation in accordance with correct identification of the sun and the moon at night. Tested in artificial conditions of illumination at night without the artificial gradient of luminance, the artificial astronomical cue is identified as the moon even when the conditions of illumination allow sun compass orientation during the day. When the artificial gradient of luminance is added, the artificial astronomical cue is identified as the sun. The role of the sky gradient of luminance in sun and moon identification is discussed on the basis of present and past findings.