• DocumentCode
    2724746
  • Title

    Retinomorphic chips that see quadruple images

  • Author

    Boahen, Kwabena

  • Author_Institution
    Penn Bioeng., Philadelphia, PA, USA
  • fYear
    1999
  • fDate
    1999
  • Firstpage
    12
  • Lastpage
    20
  • Abstract
    Retinomorphic chips may improve their spike-coding efficiency by emulating the primate retina´s parallel pathways. To this end, I recreated retinal microcircuits in a chip, Visio1, that models the four predominant ganglion-cell types. It has 104×96 photoreceptors, 4×52×48 ganglion-cells, a die size of 9.25×9.67 mm2 in 1.2 μm 5V CMOS, and consumes 11.5 mW at 5 spikes/second/neuron. Visio1 includes novel subthreshold current-mode circuits that use horizontal-cell autofeedback to decouple spatiotemporal bandpass filtering from local gain control and use amacrine-cell loop-gain modulation to adapt highpass and lowpass temporal filtering. Different ganglion cells respond to motion in a stereotyped sequence, making it possible to detect edges of one contrast or the other moving in one direction or the other. The author presents results from a multichip 2-D motion architecture, which implements Watson and Ahumada´s model of human visual-motion sensing
  • Keywords
    CMOS analogue integrated circuits; analogue processing circuits; band-pass filters; computer vision; current-mode circuits; edge detection; neural chips; 1.2 micron; 11.5 mW; 5 V; CMOS; Visio1; amacrine-cell loop-gain modulation; die size; edge detection; ganglion-cells; horizontal-cell autofeedback; local gain control; multichip 2-D motion architecture; parallel pathways; photoreceptors; predominant ganglion-cell types; quadruple images; retinal microcircuits; retinomorphic chips; spatiotemporal bandpass filtering; spike-coding efficiency; subthreshold current-mode circuits; temporal filtering; Band pass filters; Current mode circuits; Filtering; Gain control; Image edge detection; Neurons; Photoreceptors; Retina; Semiconductor device modeling; Spatiotemporal phenomena;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Microelectronics for Neural, Fuzzy and Bio-Inspired Systems, 1999. MicroNeuro '99. Proceedings of the Seventh International Conference on
  • Conference_Location
    Granada
  • Print_ISBN
    0-7695-0043-9
  • Type

    conf

  • DOI
    10.1109/MN.1999.758840
  • Filename
    758840