Hardware Implementation Of An Optimized Scaleinvariant Feature Detector For Robotic Applications

• doi: 10.1109/IST.2014.6958478
• title: Hardware implementation of an optimized scale-invariant feature detector for robotic applications
• publisher: IEEE
• isbn: 978-1-4799-5220-5
• issn: 1558-2809
• rank: 2567
• access_type: LOCKED
• content_type: Conferences
• abstract: A new architecture for the real-time detection of scale-invariant features in image sequences is presented. The system is based on a low-cost smart-camera custom board, developed to target robotic vision applications. Several optimizations of the SIFT detection procedure are proposed in order to achieve robust keypoint detection with high repeatability and recall values. As a result, a high accuracy and resource-efficient implementation of the SIFT detector is presented. The system is pipelined and streams pixel data using a 45 MHz clock, allowing keypoint detection at 150 frames per second, in video sequences with resolution 640×480. Integrating a commodity CMOS sensor, the prototype system displays keypoints at video rate, using only a fraction of the resources of a low-cost FPGA device.
• article_number: 6958478
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6958478
• html_url: https://ieeexplore.ieee.org/document/6958478/
• abstract_url: https://ieeexplore.ieee.org/document/6958478/
• publication_title: 2014 IEEE International Conference on Imaging Systems and Techniques (IST) Proceedings
• conference_location: Santorini, Greece
• conference_dates: 14-17 Oct. 2014
• publication_number: 6948378
• is_number: 6958430
• publication_year: 2014
• publication_date: 14-17 Oct. 2014
• start_page: 226
• end_page: 231
• citing_paper_count: 0
• citing_patent_count: 0
• download_count: 145
• insert_date: 20141120
• index_terms:
  • ieee_terms:
    • Filtering
    • Kernel
    • Hardware
    • Feature extraction
    • Detectors
    • Eigenvalues and eigenfunctions
    • Convolution
  • author_terms:
    • SIFT
    • computer hardware
    • real-time
    • machine vision
  • dynamic_index_terms:
    • Feature Detection
    • Robotic Applications
    • Hardware Implementation
    • Detection Procedure
    • Scale-invariant Feature Transform
    • Keypoint Detection
    • Peak Value
    • Gaussian Filter
    • Second Moment
    • Statistical Moments
    • Reference Image
    • Affine Transformation
    • Original Algorithm
    • Hessian Matrix
    • Hessian Matrices
    • Image Scale
    • Feature Matching
    • Matching Characteristics
    • Digital Signal Processing
    • Image Edge
    • Filtering Strategy
    • Filtering Scheme
    • Image Rotation
    • Filtering Stage
    • Difference Of Gaussian
    • Fixed Window Size
    • Fixed-size Window
    • Moment Matrix
    • Moment Matrices
    • Vector Modes
    • Filter Configuration
    • Tracking Applications
    • Input Image
    • Bilinear Interpolation
    • Eigenvectors
    • Power Requirements
• isbn_formats:
  • format: Electronic ISBN, value: 978-1-4799-5220-5, isbnType: New-2005
• authors:
  • Author Name: John Vourvoulakis
    Affiliation: Department of Electrical and Computer Engineering Democritus, University of Thrace, Xanthi, Greece
    Author URL: https://ieeexplore.ieee.org/author/37085509956
    ID: 37085509956
    Order: 1
    Author Affiliations:
    • Department of Electrical and Computer Engineering Democritus, University of Thrace, Xanthi, Greece
  • Author Name: John Lygouras
    Affiliation: Department of Electrical and Computer Engineering Democritus, University of Thrace, Xanthi, Greece
    Author URL: https://ieeexplore.ieee.org/author/37299296400
    ID: 37299296400
    Order: 2
    Author Affiliations:
    • Department of Electrical and Computer Engineering Democritus, University of Thrace, Xanthi, Greece
  • Author Name: John Kalomiros
    Affiliation: Department of Informatics Engineering, Technological Educational Institute of Central Macedonia, Serres, Greece
    Author URL: https://ieeexplore.ieee.org/author/37689528800
    ID: 37689528800
    Order: 3
    Author Affiliations:
    • Department of Informatics Engineering, Technological Educational Institute of Central Macedonia, Serres, Greece

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: 150 fps
• power_consumption: Approx. 650 mW (150 mW for the FPGA + 500 mW for CMOS sensor and VGA DAC)

Applications & Benefits

• cell_imaging: Not specified
• benefits: Real-time feature tracking in robotic applications.

Supporting Organizations

• supported_by: Not specified

Manuscript Details

• publication_date: 14-17 Oct. 2014

Relevancy Score

• score: 9
• missing_fields:
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • signal_to_noise_ratio
  • sensitivity
  • shutter_speed
  • noise

Processed JSON Filename

request_3cf630a4-caba-4b05-9af2-73a9573b734b-hardware_implementation_of_an_optimized_scaleinvariant_feature_detector_for_robotic_applications.json