A Computational Cmos Image Sensor Architecture Using Inpixel Pwmscibased
Mac With Reconfigurable Kernel Size
- doi: 10.1109/NewCAS58973.2024.10666327
-
title: A Computational CMOS Image Sensor Architecture
Using In-Pixel PWM-SCI-Based MAC with Reconfigurable Kernel Size
- publisher: IEEE
- isbn: 979-8-3503-6176-6
- issn: 2472-467X
- rank: 675
- access_type: LOCKED
- content_type: Conferences
-
abstract: In the era of artificial intelligence (AI)
and computer vision (CV), there is a rising demand for complementary
metaloxide semiconductor (CMOS) image sensors to process visual
information. This paper introduces a CMOS image sensor (CIS)
architecture designed to ameliorate the limitation of fixed kernel size
of on-chip image filtering by utilizing column-shared CIS circuits. We
present the overall architecture and provide details of the working
principle of the proposed CIS. The pixel and the associated
column-shared circuit are designed and simulated in a 1-poly 6-metal
0.18-µm CMOS process. The presented work paves the path for making CIS
architectures more versatile and resolves the kernel size problem in
existing solutions.
- article_number: 10666327
-
pdf_url:
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10666327
-
html_url:
https://ieeexplore.ieee.org/document/10666327/
-
abstract_url:
https://ieeexplore.ieee.org/document/10666327/
-
publication_title: 2024 22nd IEEE Interregional NEWCAS
Conference (NEWCAS)
- conference_location: Sherbrooke, QC, Canada
- conference_dates: 16-19 June 2024
- publication_number: 10666286
- is_number: 10666108
- publication_year: 2024
- publication_date: 16-19 June 2024
- start_page: 198
- end_page: 202
- citing_paper_count: 0
- citing_patent_count: 0
- download_count: 193
- insert_date: 20240917
-
index_terms:
-
ieee_terms:
- Visualization
- Simulation
- Circuits
- Neural networks
- Computer architecture
- CMOS image sensors
- Image filtering
-
author_terms:
- Artificial Intelligence
- Computer Vision
- CMOS Image Sensor
-
dynamic_index_terms:
- Image Sensor
- Camera Sensor
- Complementary Metal–oxide–semiconductor Image
- Complementary Metal Oxide Semiconductor Image
- Architectural Design
- Image Filtering
- Neural Network
- Simulation Results
- Convolution
- Image Classification
- Photodiode
- Pulse Width
- Pulsewidth Modulation
- Network Layer
- Current Source
- Current Regulations
- Frame Rate
- Analog-to-digital Converter
- Analog-to-digital
- Digital-to-analog Converter
- Photocurrent
- Intermediate Results
- Intermediate Outcomes
- Circuit Design
- Supply Voltage
- Digital Circuits
- Digital Circuitry
- Complementary Metal Oxide Semiconductor Technology
- Pixel Array
- Kernel Weight
- Objects In The Scene
- Series Of Signals
- Dimension Of The Kernel
-
isbn_formats:
-
format: Print on Demand(PoD) ISBN,
value: 979-8-3503-6176-6,
isbnType: New-2005
-
format: Electronic ISBN,
value: 979-8-3503-6175-9,
isbnType: New-2005
-
authors:
-
Author Name: Jiajun Li
Affiliation: Department of Electrical and Computer
Engineering, The University of British Columbia, Vancouver, B.C.,
Canada
Author URL:
https://ieeexplore.ieee.org/author/37985270900
ID: 37985270900
Order: 1
Author Affiliations:
-
Department of Electrical and Computer Engineering, The
University of British Columbia, Vancouver, B.C., Canada
-
Author Name: Yi Luo
Affiliation: Department of Electrical and Computer
Engineering, The University of British Columbia, Vancouver, B.C.,
Canada
Author URL:
https://ieeexplore.ieee.org/author/970306948901920
ID: 970306948901920
Order: 2
Author Affiliations:
-
Department of Electrical and Computer Engineering, The
University of British Columbia, Vancouver, B.C., Canada
-
Department of Component Development, vivo Mobile Communication
Co., Ltd., Shenzhen, China
-
Author Name: Reza Molavi
Affiliation: Department of Electrical and Computer
Engineering, The University of British Columbia, Vancouver, B.C.,
Canada
Author URL:
https://ieeexplore.ieee.org/author/37421523900
ID: 37421523900
Order: 3
Author Affiliations:
-
Department of Electrical and Computer Engineering, The
University of British Columbia, Vancouver, B.C., Canada
-
Author Name: Shahriar Mirabbasi
Affiliation: Department of Electrical and Computer
Engineering, The University of British Columbia, Vancouver, B.C.,
Canada
Author URL:
https://ieeexplore.ieee.org/author/37269465100
ID: 37269465100
Order: 4
Author Affiliations:
-
Department of Electrical and Computer Engineering, The
University of British Columbia, Vancouver, B.C., Canada
Image Sensor
- sensor_type: CMOS
- resolution: Not specified
- dynamic_range: Not specified
- pixel_size: 7 μm×7 μm
-
dark_current: 5 fA to 500 fA based on ambient light
conditions
Optical Data
- focal_length: Not specified
- aperture: Not specified
- field_of_view: Not specified
- distortion: Not specified
Performance Metrics
- frame_rate: 0-400 fps
- power_consumption: 0.8-1.4 mW per pixel
Applications & Benefits
-
cell_imaging: The sensor enables in-pixel processing
for efficient image classification and computational tasks in various
applications such as image recognition and scene analysis.
-
benefits: The proposed architecture offers a variable
kernel size for enhanced flexibility in processing, addressing the
limitations of fixed kernel sizes in existing solutions.
Supporting Organizations
-
supported_by: Natural Sciences and Engineering Research
Council of Canada (NSERC), CMC Microsystems
Manuscript Details
- publication_date: 16-19 June 2024
Relevancy Score
- score: 10
-
missing_fields:
- fill factor
- quantum efficiency
- readout speed
- noise sources
- analog-to-digital conversion techniques
- microlenses
- on-chip colour filters
- global or rolling shutters
- backside illumination
Processed JSON Filename
request_14969b5d-f21d-4b5f-b610-a8074f7ede26-a_computational_cmos_image_sensor_architecture_using_inpixel_pwmscibased_mac_with_reconfigurable_kernel_size.json