Multipole-based CSF estimates predict crack detection individual differences (2005)
The contrast sensitivity function (CSF) is a modulation transfer function estimate for the visual system within the context of a detection or discrimination model. Such estimates are often derived from measurements of the delectability of grating patches varying in spatial frequency (cycles per degree of visual angle). Such measurements are technically difficult because of the high contrast sensitivity of the visual system at some spatial frequencies. UC Berkeley scientist Stanley Klein has proposed using mulipole stimuli (edge, line, and higher spatial derivatives) to estimate CSFs. Using the Modelfest results, we find very poor agreement between CSFs estimated from gratings and multipole-like stimuli.
none
none |