• DocumentCode
    295897
  • Title

    Strategies for visual navigation, target detection and camouflage: inspirations from insect vision

  • Author

    Srinivasan, M.V.

  • Author_Institution
    Centre for Visual Sci., Australian Nat. Univ., Canberra, ACT, Australia
  • Volume
    5
  • fYear
    1995
  • fDate
    Nov/Dec 1995
  • Firstpage
    2456
  • Abstract
    Recent research, observing freely flying insects reveals a number of computational “short cuts” that insects use for perceiving their visual world in three dimensions, and navigating successfully in it. Bees segment objects from their backgrounds by sensing the apparent relative motion at the boundary between object and background. The distances to objects are gauged in terms of the apparent speeds of motion of the objects´ images, rather than by using complex stereo mechanisms to compute range. Hoverflies “shadow” conspecific by camouflaging their motion. The shadowing insect achieves this by moving in such a way as to emulate the motion of the image of a stationary object in the eye of the moving shadowy. Bees flying through a tunnel maintain equidistance to the side walls by balancing the apparent speeds of the images of the walls. Bees landing on a horizontal surface hold constant the image velocity of the surface as they approach it, thus automatically ensuring that flight speed is close to zero at touchdown. Applications of some these strategies to robot navigation and machine vision are discussed
  • Keywords
    biocybernetics; biology; computer vision; motion estimation; navigation; object recognition; vision; image motion; insect vision; machine vision; motion camouflage; robot navigation; target detection; visual navigation; Australia; Eyes; Image segmentation; Insects; Machine vision; Navigation; Object detection; Optical refraction; Robot vision systems; Shadow mapping;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks, 1995. Proceedings., IEEE International Conference on
  • Conference_Location
    Perth, WA
  • Print_ISBN
    0-7803-2768-3
  • Type

    conf

  • DOI
    10.1109/ICNN.1995.487747
  • Filename
    487747