Visual Evaluation of Computer-Generated Textures
Abstract
Textures generated by superimposing sinusoidal luminance distributions can be used to simulate the natural terrain textures often used in flight simulator imagery. Since the visual system is spatially inhomogeneous with the periphery being generally less sensitive than the center of the visual field, simpler, more easily generated textures can potentially be used to simulate terrain that is farther from the operator's point of regard. The minimal number of component sinusoids required to generate textures that are visually acceptable in the visual periphery was estimated for the discrimination of complex suprathreshold textures. Specifically, similarity ratings were obtained to determine the effects of component orientation and component phase- bandwidth on the cortical magnification factor (CMF) associated with that discrimination. The textures were designed to be both specifically by a relatively small number of localized spectral components and sufficiently complex to approximate natural images. The number of component orientations was found to be a particularly important determinant of texture discrimination in that its effect on rated similarity was largely independent of the total number of components making up the texture. When the number of components was varied, a CMF of 2 was sufficient to equate the similarity ratings obtained at 0.75 deg and 20 deg. Under the same conditions, a CMF of 4 clearly overcorrected the data. The estimated CMF for texture discrimination is much smaller than that found for the discrimination of simple 2-D spatial frequency and suggests that either quantitatively different cortical mechanisms or different cortical areas are responsible for the two types of discrimination. Gabor functions, Phase, Vision models, Orientation, Textures, Perception, Vision.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1994
- Accession Number
- ADA277201
Entities
People
- Don R. Lyon
- George A. Geri
- Yehoshua Y. Zeevi
Organizations
- University of Dayton