On Recent Results In Demosaicing Of Samsung 108Mp Cmos Sensor Using Deep Learning

• doi: 10.1109/TENSYMP52854.2021.9550945
• title: On recent results in demosaicing of Samsung 108MP CMOS sensor using deep learning
• publisher: IEEE
• isbn: 978-1-6654-0027-5
• issn: 2640-821X
• rank: 3170
• access_type: LOCKED
• content_type: Conferences
• abstract: There is a recent trend change in using non-Bayer Color Filter Array (CFA) patterns, allowing superior image quality in low-light conditions. However, these new CFA patterns have weak point on full image reconstruction showing severe visual artifacts and low details reconstruction after demosaicing. In this work, we address aforementioned problems by using deep learning approach for new CFA type - Nonacell or Nonapixel, introduced by Samsung 108MP HMX CMOS image sensor. Experimental results show that proposed method not only allows suppression of visual artifacts and perfect details and edge restoration, but also shows superior objective image quality, exceeding 40dB in CPSNR for popular Kodak dataset. Finally, our method is computationally efficient due to network structure and feasible for on-device deployment after optimization.
• article_number: 9550945
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9550945
• html_url: https://ieeexplore.ieee.org/document/9550945/
• abstract_url: https://ieeexplore.ieee.org/document/9550945/
• publication_title: 2021 IEEE Region 10 Symposium (TENSYMP)
• conference_location: Jeju, Korea, Republic of
• conference_dates: 23-25 Aug. 2021
• publication_number: 9550759
• is_number: 9550761
• publication_year: 2021
• publication_date: 23-25 Aug. 2021
• start_page: 1
• end_page: 4
• citing_paper_count: 4
• citing_patent_count: 0
• download_count: 944
• insert_date: 20211004
• index_terms:
  • ieee_terms:
    • Image quality
    • Deep learning
    • Visualization
    • Image color analysis
    • Image edge detection
    • Image restoration
    • Image reconstruction
  • author_terms:
    • CMOS sensor
    • Tetracell
    • Nanocell
    • Nanopixel
    • color filter array
    • image restoration
    • demosacing
    • deep learning
    • convolutional neural network
  • dynamic_index_terms:
    • Image Quality
    • Superior Quality
    • Superior Characteristics
    • Superior Features
    • Matrix Form
    • Downscaling
    • Color Space
    • Colour Space
    • Color Components
    • Level Of Resolution
    • Subjective Quality
    • Discrete Fourier Transform
    • Objective Quality
    • Improve Image Quality
    • Detailed Reconstruction
    • Image Quality Evaluation
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-6654-0027-5, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-6654-0026-8, isbnType: New-2005
• authors:
  • Author Name: Irina Kim
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37088709191
    ID: 37088709191
    Order: 1
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
  • Author Name: Dongpan Lim
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37086548504
    ID: 37086548504
    Order: 2
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
  • Author Name: Youngil Seo
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37088986573
    ID: 37088986573
    Order: 3
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
  • Author Name: Jeongguk Lee
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37088484048
    ID: 37088484048
    Order: 4
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
  • Author Name: Yunseok Choi
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37089269701
    ID: 37089269701
    Order: 5
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
  • Author Name: Seongwook Song
    Affiliation: Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37089259034
    ID: 37089259034
    Order: 6
    Author Affiliations:
    • Sensor Business, Device Solutions, Samsung Electronics, Hwaseong, South Korea

Image Sensor

• sensor_type: CMOS
• resolution: 108MP
• dynamic_range: 40dB
• 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

Applications & Benefits

• cell_imaging: not provided
• benefits: The method enables superior image quality in low-light conditions and reduces visual artifacts.

Supporting Organizations

• supported_by: Samsung Electronics

Manuscript Details

• publication_date: 23-25 Aug. 2021

Relevancy Score

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

Processed JSON Filename

request_0c936213-5672-4900-8af7-f84c2306fb40-on_recent_results_in_demosaicing_of_samsung_108mp_cmos_sensor_using_deep_learning.json