Lightweight Object Detection Model For A Cmos Image Sensor With Binary Feature Extraction

• doi: 10.1109/SENSORS60989.2024.10784814
• title: Lightweight Object Detection Model for a CMOS Image Sensor with Binary Feature Extraction
• publisher: IEEE
• isbn: 979-8-3503-6352-4
• issn: 1930-0395
• rank: 2877
• access_type: LOCKED
• content_type: Conferences
• abstract: For the coming IoT age, we propose an object detection system using a CMOS image sensor capable of extracting binary feature data in order to reduce the power consumption of recognition systems. First, a lightweight deep neural network (DNN) for feature data is verified based on YOLOv7. The presented model is comparable to the YOLOv7-tiny in the number of parameters and FLOPs, but improves the object recognition accuracy of large objects APL 50 by 6.6%. Additionally, the presented model for the feature data reduces GPU power consumption by 39.2% compared to the original YOLOv7 that processes RGB color images. Secondly, an on-chip signal processing technique for the CMOS image sensor is proposed to extract binary feature data. The simulation results demonstrate that the size of 1-bit feature data is reduced by 96.0% at the cost of a image classification accuracy degradation of 8.3% compared to the 1-bit RGB color image.
• article_number: 10784814
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10784814
• html_url: https://ieeexplore.ieee.org/document/10784814/
• abstract_url: https://ieeexplore.ieee.org/document/10784814/
• publication_title: 2024 IEEE SENSORS
• conference_location: Kobe, Japan
• conference_dates: 20-23 Oct. 2024
• publication_number: 10783834
• is_number: 10784457
• publication_year: 2024
• publication_date: 20-23 Oct. 2024
• start_page: 1
• end_page: 4
• citing_paper_count: 0
• citing_patent_count: 0
• download_count: 77
• insert_date: 20241217
• index_terms:
  • ieee_terms:
    • Semiconductor device modeling
    • Power demand
    • Accuracy
    • Color
    • Object detection
    • Artificial neural networks
    • CMOS image sensors
    • Feature extraction
    • Data models
    • Data mining
  • author_terms:
    • CMOS image sensor
    • feature extraction
    • object detection
    • data reduction
    • run-length encoding
  • dynamic_index_terms:
    • Object Detection
    • Image Sensor
    • Camera Sensor
    • Binary Feature Extraction
    • Deep Neural Network
    • Power Consumption
    • Binary Data
    • Binary Values
    • Degradation Cost
    • RGB Color Images
    • Amount Of Information
    • Convolutional Layers
    • Intersection Over Union
    • Jaccard Similarity
    • Jaccard Distance
    • Jaccard Index
    • Pixel Array
    • Conventional Sensors
    • Lightweight Model
    • Channel Reduction
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 979-8-3503-6352-4, isbnType: New-2005
  • format: Electronic ISBN, value: 979-8-3503-6351-7, isbnType: New-2005
• authors:
  • Author Name: Keiichiro Kuroda
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/105127314267843
    ID: 105127314267843
    Order: 1
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
  • Author Name: Yudai Morikaku
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/160909248169271
    ID: 160909248169271
    Order: 2
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
  • Author Name: Yu Osuka
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/802057822153687
    ID: 802057822153687
    Order: 3
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
  • Author Name: Ryoya Iegaki
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/845043578389369
    ID: 845043578389369
    Order: 4
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
  • Author Name: Kota Yoshida
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/37086859376
    ID: 37086859376
    Order: 5
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
  • Author Name: Shunsuke Okura
    Affiliation: Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan
    Author URL: https://ieeexplore.ieee.org/author/37089734086
    ID: 37089734086
    Order: 6
    Author Affiliations:
    • Research Organization of Science and Engineering Ritsumeikan University, Shiga, Japan

Image Sensor

• sensor_type: CMOS
• resolution: Not specified
• 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

• power_consumption: Reduced power consumption by 39.2% compared to original YOLOv7
• noise: Electrons (e-) not specified

Applications & Benefits

• cell_imaging: Not specified
• benefits: Reduction in power consumption and increased object detection accuracy

Supporting Organizations

• supported_by: Ritsumeikan University

Manuscript Details

• publication_date: 20-23 Oct. 2024

Relevancy Score

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

Processed JSON Filename

request_14a30c53-4efc-474b-a37f-dbad52d18db4-lightweight_object_detection_model_for_a_cmos_image_sensor_with_binary_feature_extraction.json