Journal of the Optical Society of America最新文献

We investigate some limits to phase processing in the human visual system with two-dimensional textured images. Results indicate that, although phase sensitivity increases with contrast and energy components of the image, observers cannot discriminate between images and their 45 degrees phase-quantized versions under brief exposure and lower- (less than or equal to 30%) contrast conditions. These results seem to be frequency independent although modulated by different energy levels at each two-dimensional frequency band. Finally, we have analyzed some past texture-discrimination results that occur under identical amplitude-spectra conditions. Here also phase-quantization differences seem to constitute an adequate explanation of discrimination performance.

We have studied the ability of observers to discriminate between suprathreshold vertical sinusoidal spatial-frequency gratings on the basis of spatial frequency. The results show that spatial-frequency discrimination is not a smooth function of spatial frequency but instead appears regularly segmented. Similar results were also obtained in an experiment in which observers discriminated between pairs of narrow vertical lines on the basis of their separation. Angular resolutions achieved for both discrimination tasks were less than the spacing between photo-receptors, requiring some type of neural interpolation. The similarity between the two sets of data indicates that discrimination between spatial-frequency gratings is probably based on the separation between two features exactly one cycle apart. We suggest that the segmentation reflects the existence of neural-image representations with discrete levels of spatial accuracy.

With a modified step-by-step brightness-matching procedure, a series of colors, with dominant wavelengths from 400 to 670 nm, was adjusted so that the saturations and brightnesses of the colors appeared equal to those of the reference, which was a mixture of 570-nm and white light. The results show that equally bright and equally saturated colors are not equal in luminance. We also report a saturation function of spectral lights derived by utilizing these equally bright and equally saturated colors. Finally, our equally saturated colors do not plot as a circle in the 1976 CIE u', v' space, which indicates some limitations of this uniform chromaticity diagram.

In the measurements of contrast thresholds for stationary, stabilized images, the choice of the psychophysical procedure may profoundly affect the results by constraining the time course of the stimulus presentation. With some procedures, the subject is unable to ignore transients; as a result, he necessarily responds to onset and offset transients rather than to the intended stimulus. Here we report significant effects of (a) the rapidly of onset and (b) the duration of the stimulus on thresholds for stabilized sinusoidal-grating patterns; these effects occur with both forced-choice and standard yes/no procedures. No value of either parameter (a) or (b) can eliminate such presentation effects when these procedures are used. However, we have devised a new, continuous-presentation procedure that yields results that do not depend on these parameters. Contrast thresholds for stabilized and for unstabilized stimuli measured by this procedure differ by 1 log unit or more, in good agreement with the results obtained by careful use of the method of adjustment.

The brightness of lights defined by heterochromatic flicker photometry as being of equal luminance was determined by direct comparison with a white standard. Stimuli included mixtures of white and colored lights. Eight different colorimetric purities were tested for each of eleven primaries. Adaptation to a known white was maintained. Resulting brightnesses were expressed in terms of the ratio of the amount of light required for a brightness match to the amount required for a flicker photometric match (B:F). B:F ratios varied with the primary, being largest for red and blue primaries. The variation of B:F ratio with colorimetric purity differed depending on the primary, although for any given dominant wavelength the highest-purity stimulus always had the largest B:F ratio. The data qualitatively exclude a number of simple models of brightness generation.

The negative afterimage from grating can be considered equivalent to a renal grating, the contrast of which decreases over time. The interval between the onset of the afterimage and the time at which the effective contrast of the afterimage falls below threshold defines afterimage duration. In a series of experiments with several predictions based on this formulation were confirmed. Square-wave gratings yielded longer afterimage durations than sinusoidal gratings, a difference that is attributable to the difference in threshold between these two types of grating. Also, grating adaptation before afterimage induction was found to abbreviate afterimage duration because of threshold elevation. Finally, it was found that, even after fading to invisibility, an afterimage could interact with a real grafting to influence threshold performance on a forced-choice detection task.