Pixelbased Pipeline Hardware Architecture For Highperformance Haarlike Feature Extraction

• doi: 10.1109/APCCAS.2016.7804044
• title: Pixel-based pipeline hardware architecture for high-performance Haar-like feature extraction
• publisher: IEEE
• isbn: 978-1-5090-1571-9
• issn:
• rank: 3274
• access_type: LOCKED
• content_type: Conferences
• abstract: Feature extraction, which is one of the basic tasks for pattern recognition, has often high computational cost and large memory usage. In this work, we propose a pixel-based pipeline hardware architecture for Haar-like feature extraction, implemented in 0.18 μm CMOS technology with 1.76 mm2 core area. Pixel-input speed relies on the working frequency of the image sensor so that features are extracted in real time without on-chip image buffer and complex computational procedures. The fabricated chip consumes 4.78 mW power at 1.8 V supply voltage and 12.5 MHz frequency during 30 fps VGA video input. Furthermore, a processing time of 3.07 ms per VGA frame with power dissipation of 36.25 mW at 100 MHz frequency is possible.
• article_number: 7804044
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7804044
• html_url: https://ieeexplore.ieee.org/document/7804044/
• abstract_url: https://ieeexplore.ieee.org/document/7804044/
• publication_title: 2016 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS)
• conference_location: Jeju, Korea (South)
• conference_dates: 25-28 Oct. 2016
• publication_number: 7786273
• is_number: 7803879
• publication_year: 2016
• publication_date: 25-28 Oct. 2016
• start_page: 611
• end_page: 612
• citing_paper_count: 0
• citing_patent_count: 0
• download_count: 105
• insert_date: 20170105
• index_terms:
  • ieee_terms:
    • Computer architecture
    • Feature extraction
    • Pipelines
    • Hardware
    • Microprocessors
    • Time-frequency analysis
    • Power demand
  • author_terms:
    • feature extraction
    • pixel-based pipeline
    • Haar-like feature
  • dynamic_index_terms:
    • Hardware Architecture
    • Haar-like Features
    • Haar Cascade
    • Architecture For Feature Extraction
    • Processing Time
    • Image Sensor
    • Camera Sensor
    • Supply Voltage
    • MHz Frequency
    • Power Consumption
    • Processing Speed
    • Psychomotor Speed
    • Detection Window
    • Clock Cycles
    • Scale-invariant Feature Transform
    • Speeded Up Robust Features
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-5090-1571-9, isbnType: New-2005
  • format: USB ISBN, value: 978-1-5090-1569-6, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-5090-1570-2, isbnType: New-2005
• authors:
  • Author Name: Yuki Fujita
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/37086108161
    ID: 37086108161
    Order: 1
    Author Affiliations:
    • Hiroshima University, Japan
  • Author Name: Fengwei An
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/38235502500
    ID: 38235502500
    Order: 2
    Author Affiliations:
    • Hiroshima University, Japan
  • Author Name: Aiwen Luo
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/37086102073
    ID: 37086102073
    Order: 3
    Author Affiliations:
    • Hiroshima University, Japan
  • Author Name: Xiangyu Zhang
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/37085836690
    ID: 37085836690
    Order: 4
    Author Affiliations:
    • Hiroshima University, Japan
  • Author Name: Lei Chen
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/37085589660
    ID: 37085589660
    Order: 5
    Author Affiliations:
    • Hiroshima University, Japan
  • Author Name: Hans Jürgen Mattausch
    Affiliation: Hiroshima University, Japan
    Author URL: https://ieeexplore.ieee.org/author/37269154600
    ID: 37269154600
    Order: 6
    Author Affiliations:
    • Hiroshima University, Japan

Image Sensor

• sensor_type: CMOS
• resolution: 640x480
• dynamic_range: not specified
• pixel_size: not specified
• dark_current: not specified

Optical Data

• focal_length: not specified
• aperture: not specified
• field_of_view: not specified
• distortion: not specified

Performance Metrics

• frame_rate: 30 fps
• power_consumption: 4.78 mW

Applications & Benefits

• cell_imaging: Not specifically mentioned in the paper.
• benefits: High energy efficiency and low power consumption.

Supporting Organizations

• supported_by: Japan's Ministry of Science and Education, University of Tokyo, Rohm, Synopsys, Cadence.

Manuscript Details

• publication_date: 25-28 Oct. 2016

Relevancy Score

• score: 8
• missing_fields:
  • fill factor
  • quantum efficiency
  • readout speed
  • noise sources
  • analog-to-digital conversion techniques
  • microlenses
  • on-chip color filters
  • global or rolling shutters
  • backside illumination

Processed JSON Filename

request_faad2e3f-8512-4639-854d-b4cbf43b4fe2-pixelbased_pipeline_hardware_architecture_for_highperformance_haarlike_feature_extraction.json