Edge raggedness is a psychophysical term which refers to the fidelity of a high-contrast edge, which has been degraded by some process, compared to an ideal edge. It has been suggested that the processing of high-contrast edges shares the same spatial-frequency processing mechanisms with that of sinewave gratings. This study uses first-order psychophysical principles, determined using sinewave grating stimuli, in a model to process edges in order to derive a quantitative metric proportional to perceived edge raggedness. The derived metric is the rms power of the processed edge power spectrum. First, the need for a two-dimensional analysis is established, then a model is developed which predicts the results of a psychophysical raggedness experiment. Finally, several distorted edge types are processed in order to determine edge raggedness. The two-dimensional single channel approach is successful in predicting the relative raggedness, as a function of specific distortion parameters, for various distorted edge types. Further psychophysical testing is needed to establish the exact relationship between perceived raggedness and the rms power metric.
Library of Congress Subject Headings
Visual perception--Mathematical models; Physiological optics
Department, Program, or Center
Chester F. Carlson Center for Imaging Science (COS)
Mannix, Michael G., "The prediction of edge raggedness via a single-channel filter model of the visual system" (1987). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus