• DocumentCode
    3402047
  • Title

    High speed CMOS vision chips

  • Author

    Nanjian Wu

  • Author_Institution
    State Key Lab. for Super Lattices & Microstructures, Chinese Acad. of Sci., Beijing, China
  • fYear
    2011
  • fDate
    7-10 Aug. 2011
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    This paper presents novel high speed vision chips based on multiple levels of parallel processors. The chip integrates CMOS image sensor, multiple-levels of SIMD parallel processors and an embedded microprocessor unit. The multiple-levels of SIMD parallel processors consist of an array processor of SIMD processing elements (PEs) and a column of SIMD row processors (RPs). The PE array and RPs have an O(N×N) parallelism and an O(N) parallelism, respectively. The PE array, RPs and MPU can execute low-, mid- and high-level image processing algorithms, respectively. Prototype chips are fabricated using the 0.18μm CMOS process. Applications including target tracking, pattern extraction and image recognition are demonstrated.
  • Keywords
    CMOS image sensors; image processing; image recognition; parallel processing; CMOS image sensor; O(N×N) parallelism; O(N) parallelism; SIMD parallel processors; SIMD processing element array processor; SIMD row processors; embedded microprocessor unit; high speed CMOS vision chips; high-level image processing algorithm; image recognition; low-level image processing algorithm; mid-level image processing algorithm; parallel processor multiple levels; pattern extraction; size 0.18 mum; Complexity theory; Image recognition; Morphology; Parallel processing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems (MWSCAS), 2011 IEEE 54th International Midwest Symposium on
  • Conference_Location
    Seoul
  • ISSN
    1548-3746
  • Print_ISBN
    978-1-61284-856-3
  • Electronic_ISBN
    1548-3746
  • Type

    conf

  • DOI
    10.1109/MWSCAS.2011.6026285
  • Filename
    6026285