• DocumentCode
    254461
  • Title

    A second-generation noise-immune motion detection image sensor for moving object tracking application

  • Author

    Xiangyu Zhang ; Shoushun Chen

  • fYear
    2014
  • fDate
    10-12 Dec. 2014
  • Firstpage
    252
  • Lastpage
    255
  • Abstract
    In this paper, we present a noise-immune high-speed motion detection CMOS image sensor. The proposed architecture is quite appealing for motion object tracking applications as it can detect motion with a high immunity to random noise. Our pixel has a front-end event generator to detect and convert temporal light intensity changes into binary events. By on-chip implementing an event-clustering algorithm using a mixed-mode circuit to process these events, our proposed sensor can remove random discrete noise events while report only active motion events to the off-chip post imaging processors. The proposed testing structure was implemented with a 5×5 pixel array in a standard 65-nm CMOS process. Each pixel occupies an area of 35×35 μm2 with a fill factor of ~15%. Power consumption for each pixel is ~5.4 μw under 1.8 V power supply.
  • Keywords
    CMOS image sensors; image denoising; image motion analysis; integrated circuit noise; mixed analogue-digital integrated circuits; object tracking; pattern clustering; random noise; sensor arrays; event-clustering algorithm; front-end event generator; mixed-mode circuit; moving object tracking application; off-chip post imaging processor; random discrete noise event removal; second-generation noise-immune motion detection CMOS image sensor; size 65 nm; temporal light intensity conversion; temporal light intensity detection; Arrays; Capacitors; Clustering algorithms; Couplings; Image sensors; Motion detection; Noise;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Integrated Circuits (ISIC), 2014 14th International Symposium on
  • Conference_Location
    Singapore
  • Type

    conf

  • DOI
    10.1109/ISICIR.2014.7029462
  • Filename
    7029462