An Intrascene Wide Dynamic Range Cmos Image Sensor Using Dualgain Ramp Generator For Machine Vision Applications

doi: 10.1109/TIM.2024.3440392
title: An Intrascene Wide Dynamic Range CMOS Image Sensor Using Dual-Gain Ramp Generator for Machine Vision Applications
publisher: IEEE
isbn:
issn: 1557-9662
rank: 1517
access_type: LOCKED
content_type: Journals
abstract: This study presents an advanced dual-imaging CMOS image sensor (CIS) optimized for intrascene wide dynamic range (WDR) machine vision applications. Based on examining the operational characteristic and structure of traditional ramp signal generator used in CISs, we proposed a new dual-ramp signal generator, enabling the concurrent generation of ramp signals with diverse analog gains (AGs). By employing a top and bottom readout array configuration, the proposed CIS enables the simultaneous conversion of pixel data with differing AGs using the proposed dual-ramp signal generator. This approach enables real-time dual imaging without any modification of standard pixel structures, effectively accommodating varying light intensities while avoiding motion artifacts. To further enhance the image quality, this work presents an efficient two-step pre-emphasis ramping technique. It effectively minimizes the column fixed pattern noise (CFPN) that typically arises due to the inherent gain changes in ramp generators. As a result, this dual-imaging CIS can serve as a promising solution for effective intrascene WDR imaging for machine vision. This prototype CIS with the proposed edge detection counter was fabricated using a 0.11- $\mu $ m CIS process with a pixel resolution of 1280 (H) $\times 720$ (V). The prototype CIS operates with a power consumption of 6.31 mW and achieves a frame rate of 59 frames per second (fps). The prototype CIS demonstrated the figures of merits (FoMs) of 116.1 pJ/frame/pixel, $36.04{\%}\cdot \text {fJ}$ /frame/pixel, and $0.035{\%}\cdot \text {fJ}$ /frame/pixel/ $2^{\text {N}}$ .
article_number: 10630871
pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10630871
html_url: https://ieeexplore.ieee.org/document/10630871/
abstract_url: https://ieeexplore.ieee.org/document/10630871/
publication_title: IEEE Transactions on Instrumentation and Measurement
conference_location:
conference_dates:
publication_number: 19
is_number: 10367905
publication_year: 2024
publication_date: 2024
start_page: 1
end_page: 8
citing_paper_count: 0
citing_patent_count: 0
download_count: 514
insert_date: 20240808
index_terms:
- ieee_terms:
  - Generators
  - Imaging
  - Analog-digital conversion
  - Resistors
  - Noise
  - Signal generators
  - Prototypes
- author_terms:
  - Analog gain (AG)
  - CMOS imager sensor (CIS)
  - column fixed pattern noise (CFPN)
  - dual-gain ramp signal generator
  - dual-imaging CIS
  - intrascene wide dynamic range (WDR)
  - two-step pre-emphasis
- dynamic_index_terms:
  - Dynamic Range
  - Image Sensor
  - Camera Sensor
  - Machine Vision
  - Visual Navigation
  - Machine Vision Applications
  - Ramp Generator
  - Image Quality
  - Signal Propagation
  - Signal Generator
  - Figure Of Merit
  - Changes In Gain
  - Dual Imaging
  - Pixel Data
  - Data For Each Pixel
  - Two-step Technique
  - Light Conditions
  - Performance Metrics
  - Control Signal
  - Object Detection
  - Object Recognition
  - Gain Control
  - Image Capture
  - Bottom Side
  - Dynamic Vision Sensor
  - Event Camera
  - Motion Blur
  - Low Light Conditions
  - High-speed Imaging
  - Pixel Array
  - Accurate Object Detection
  - Seamless Integration
authors:
- Author Name: Hyeon-June Kim
  Affiliation: Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea
  Author URL: https://ieeexplore.ieee.org/author/37085397854
  ID: 37085397854
  Order: 1
  Author Affiliations:
  - Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea
- Author Name: Soon-Kyu Kwon
  Affiliation: Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea
  Author URL: https://ieeexplore.ieee.org/author/353117330166274
  ID: 353117330166274
  Order: 2
  Author Affiliations:
  - Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea
- Author Name: Tae-Hoon Eom
  Affiliation: Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea
  Author URL: https://ieeexplore.ieee.org/author/729879972743117
  ID: 729879972743117
  Order: 3
  Author Affiliations:
  - Department of Semiconductor Engineering, Seoul National University of Science and Technology, Seoul, South Korea

Image Sensor

sensor_type: CMOS
resolution: 1280 (H) × 720 (V)
dynamic_range: 66 dB
pixel_size: 4 µm
dark_current: not provided

Optical Data

focal_length: not provided
aperture: not provided
field_of_view: not provided
distortion: not provided

Performance Metrics

frame_rate: 59 fps
signal_to_noise_ratio: not provided
sensitivity: not provided
shutter_speed: not provided
power_consumption: 6.31 mW
noise: 0.31 mV rms (AG 1), 0.53 mV rms (AG 2), 0.92 mV rms (AG 4), 1.62 mV rms (AG 8), 2.89 mV rms (AG 16)

Applications & Benefits

cell_imaging: used for machine vision applications, especially for high-speed object detection and wide dynamic range
benefits: Ability to process images with different analog gains simultaneously, improved image quality with reduced motion blur, effective in low and variable light conditions

Supporting Organizations

supported_by: National Research Foundation of Korea (NRF); Ministry of Science and Information and Communications Technology (ICT)

Manuscript Details

publication_date: 2024

Relevancy Score

score: 10

Processed JSON Filename

request_07e72fef-198f-42b3-9c2c-58c1fc91e9df-an_intrascene_wide_dynamic_range_cmos_image_sensor_using_dualgain_ramp_generator_for_machine_vision_applications.json