Original Abstract:
Computational models and occupational vision standards for acuity and
contrast sensitivity
A J Ahumada, B L Beard
NASA Ames Research Center, NASA Mail Stop 262-2, Moffett Field, CA 94035, USA
e-mail: al.ahumada@nasa.gov
website: http://vision.arc.nasa.gov/personnel/al/talks/03ecvp/
Using aircraft maintenance as an example, we describe a way of using image discrimination models to assist in the development of performance-related occupational vision standards. First, we calibrate a simple image discrimination model so that it accurately predicts a subset of the ModelFest stimuli that are most similar to the original and blurred images of interest, airframe, and powerplant cracks. We then predict the visibility of simulated cracks varying in length and width viewed by observers having various visual acuities and contrast sensitivities. Poor visual acuity is simulated by increased Gaussian blur of the images. Poor contrast sensitivity is simulated by broadband reduction in contrast. The simulation results allow performance requirements described in terms of physical images to be translated into acuity and contrast sensitivity standards. Variables such as age may be brought into this framework through previous studies relating those variables to the contrast sensitivity function.