Purchase this article with an account.
R. Rattan, J. Rovamo, J.M. Wild; Masking Effects and Spatial Limitations of White Pixelated Noise . Invest. Ophthalmol. Vis. Sci. 2006;47(13):5355.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The spectral density of spatio–temporal noise can be modified by changing the root mean square (rms) contrast, the noise check size and the duration of the noise check. The latter two parameters can be increased to strengthen a noise mask up to a point, beyond which it can no longer be considered to mimic ‘white’ noise. We compared the spatial cut–off, in terms of noise checks per cycle, when masking stationary and drifting cosine gratings.
The authors measured contrast energy thresholds, E, for stationary and drifting (16 Hz) vertical cosine gratings, at spatial frequencies of 1 and 8 cycles per degree. The gratings were embedded in spatio–temporal noise and contrast energy thresholds were determined against a range of noise check sizes. The noise spectral density of spatio–temporal noise was calculated by multiplying the duration of the noise frame, the area of the noise check, and the noise rms contrast squared.
With fewer noise checks per grating cycle, the contrast energy thresholds increased as the spectral density of noise strengthened. The signal–to–noise ratio remained constant until the grating contrast threshold started to decrease indicating that the noise was no longer effective as a mask. The stationary gratings of 1 and 8 cycles per degree were masked down to 3.6 and 3.5 noise checks per cycle, respectively. The drifting gratings of 1 and 8 cycles per degree were masked down to 2.4 and 2.0 checks per cycle, respectively.
The spatial cut–off for a two–dimensional, spatio–temporal mask is dependent on the temporal parameters of the signal. The critical numbers of checks per cycle that will mimic ‘white’ noise are shown to be smaller for drifting gratings than for stationary gratings.
This PDF is available to Subscribers Only