Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Demultiplexing the hue and luminance signals in r-g X-cells

Open Access Open Access

Abstract

The primate fovea enjoys excellent color and spatial vision despite being dominated by a cell that is specialized for neither color nor acuity. Ingling and Martinez [Vision Res. 23, 1495 (1983)] have shown that the r-g X-cell adds surround and center responses for color and subtracts surround from center responses for acuity. In effect, hue signals are encoded by a low pass filter and luminance (form) signals are encoded by a bandpass filter. The multiplexed signals are transmitted to the cortex, where a demultiplexing operation must take place. Two demultiplexing algorithms—Russell (nonlinear) filtering and cancellation-have been discussed in the literature. We introduce a linear filtering algorithm that can be implemented easily in neural systems. Computational experiments with Mondrians show that linear filtering with a thresholding operation produces results identical to Russell filtering. We discuss psychophysical and electrophysiological evidence for each method and show how cortical color opponent, double opponent, and achromatic (high resolution) cells can arise from the simple opponent r-g type I cells. The multiplexing/demultiplexing operation allows the visual system to substitute computation for bandwidth in the r-g X-cell channel.

© 1989 Optical Society of America

PDF Article
More Like This
Benham's Top: an information theoretic approach

Vincent A. Billock
THMM4 OSA Annual Meeting (FIO) 1989

Temporal tuning of Fechner–Benham subjective color

Vincent A. Billock
ThOO7 OSA Annual Meeting (FIO) 1990

Computer model of foveal bipolar cells in human retina: spectral responses and chromatic adaptation

R. Siminoff and T. H. Harding
TUU32 OSA Annual Meeting (FIO) 1989

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.