Jove
Visualize
Contact Us

Related Experiment Videos

Physical limits of acuity and hyperacuity.

W S Geisler

    Journal of the Optical Society of America. A, Optics and Image Science
    |July 1, 1984
    PubMed
    Summary
    This summary is machine-generated.

    This study derives an ideal detector for visual tasks, limited by physical factors. It reveals that vernier acuity is significantly better than two-point resolution, offering insights into optimal visual perception.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Image quality assessment based on a degradation model.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
    Same author

    Motion direction signals in the primary visual cortex of cat and monkey.

    Visual neuroscience·2002
    Same author

    Edge co-occurrence in natural images predicts contour grouping performance.

    Vision research·2001
    Same author

    Spikes versus BOLD: what does neuroimaging tell us about neuronal activity?

    Nature neuroscience·2000
    Same author

    Motion streaks provide a spatial code for motion direction.

    Nature·1999
    Same author

    Visual cortex neurons in monkeys and cats: detection, discrimination, and identification.

    Visual neuroscience·1997
    JoVE
    x logofacebook logolinkedin logoyoutube logo
    ABOUT JoVE
    OverviewLeadershipBlogJoVE Help Center
    AUTHORS
    Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
    LIBRARIANS
    TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
    RESEARCH
    JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
    EDUCATION
    JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
    Terms & Conditions of Use
    Privacy Policy
    Policies

    Area of Science:

    • Vision science
    • Neuroscience
    • Image processing

    Background:

    • The performance of visual systems is constrained by physical limitations.
    • Understanding ideal detector models is crucial for evaluating biological vision.

    Purpose of the Study:

    • To derive an ideal detector model for visual discrimination tasks.
    • To predict performance limits based on quantal fluctuations, optics, and retinal mosaic.
    • To compare the performance of ideal and real (human) visual detectors.

    Main Methods:

    • Derivation of an ideal detector for the two-alternative forced-choice paradigm.
    • Analysis of performance limitations including quantal fluctuations and retinal receptor properties.
    • Modeling of two-point acuity and hyperacuity tasks.

    Related Experiment Videos

    Main Results:

    • The ideal detector's two-point resolution is ~10x worse than its vernier acuity or separation discrimination.
    • Two-point resolution scales with spot intensity to the -1/4 power.
    • Vernier acuity and separation discrimination scale with spot intensity to the -1/2 power.
    • Optimal receptive field shapes for arbitrary stimuli can be determined.

    Conclusions:

    • Ideal visual detection is limited by fundamental physical constraints.
    • Vernier acuity demonstrates superior performance compared to simple resolution tasks.
    • The derived model provides a framework for understanding optimal visual processing and designing artificial systems.