A Low-Power Compression-Based CMOS Image Sensor With Microshift-Guided SAR ADC

Xiaopeng Zhong, Bo Zhang, Amine Bermak, Chi Ying Tsui, Man Kay Law

Research output: Contribution to journalArticle

Abstract

This brief presents a low-power compression-based CMOS image sensor for wireless vision applications. The sensor implements low-bit-depth imaging with planned sensor distortion (PSD) to effectively compress both data bandwidth and processing power while maintaining high reconstruction image quality. Accordingly, a column-parallel microshift-guided successive-approximation-register (SAR) ADC is proposed to enable 3-bit PSD imaging based on a 3×3 pattern. To support normal imaging with low area overhead, the circuit is reconfigurable as an 8-bit SAR/Single-Slope (SS) ADC. The data bandwidth is further compressed by a customized lossless encoder based on predictive coding and run length coding. A 256×216 prototype imaging system composed of the compression-based image sensor and the lossless encoder is fabricated in a 0.18 μm standard CMOS process. Measurement results show that the image sensor achieves 3 bit/pixel (bpp) with 34.9 dB reconstruction peak signal-to-noise ratio (PSNR) and 0.91 structural similarity index (SSIM). With 1/4 spatial downsampling and lossless encoding, 0.31 bpp is obtained with 29.2 dB PSNR and 0.83 SSIM. The sensor consumes 14.8 μW (full resolution) and 4.3 μW (downsampling) at 15 fps, achieving state-of-the-art FoMs of 17.8 and 5.2 pJ/pixel∙frame respectively. Including the encoder, the overall system dissipates as low as 1.2 μJ/frame, making it an attractive solution for wireless sensor networks.

Original languageEnglish
JournalIEEE Transactions on Circuits and Systems II: Express Briefs
DOIs
Publication statusAccepted/In press - 23 Jul 2018

Fingerprint

Image sensors
Sensors
Imaging techniques
Signal to noise ratio
Pixels
Bandwidth
Imaging systems
Image quality
Wireless sensor networks
Networks (circuits)
Processing

Keywords

  • Bandwidth
  • CMOS image sensor
  • focal-plane compression
  • Image coding
  • Image quality
  • Image reconstruction
  • Image sensors
  • low bit depth
  • microshift
  • planned sensor distortion (PSD)
  • SAR ADC
  • Wireless sensor networks
  • wireless sensor networks.

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

A Low-Power Compression-Based CMOS Image Sensor With Microshift-Guided SAR ADC. / Zhong, Xiaopeng; Zhang, Bo; Bermak, Amine; Tsui, Chi Ying; Law, Man Kay.

In: IEEE Transactions on Circuits and Systems II: Express Briefs, 23.07.2018.

Research output: Contribution to journalArticle

@article{64b0e6424d8a4e87bbd5a694324d1403,
title = "A Low-Power Compression-Based CMOS Image Sensor With Microshift-Guided SAR ADC",
abstract = "This brief presents a low-power compression-based CMOS image sensor for wireless vision applications. The sensor implements low-bit-depth imaging with planned sensor distortion (PSD) to effectively compress both data bandwidth and processing power while maintaining high reconstruction image quality. Accordingly, a column-parallel microshift-guided successive-approximation-register (SAR) ADC is proposed to enable 3-bit PSD imaging based on a 3×3 pattern. To support normal imaging with low area overhead, the circuit is reconfigurable as an 8-bit SAR/Single-Slope (SS) ADC. The data bandwidth is further compressed by a customized lossless encoder based on predictive coding and run length coding. A 256×216 prototype imaging system composed of the compression-based image sensor and the lossless encoder is fabricated in a 0.18 μm standard CMOS process. Measurement results show that the image sensor achieves 3 bit/pixel (bpp) with 34.9 dB reconstruction peak signal-to-noise ratio (PSNR) and 0.91 structural similarity index (SSIM). With 1/4 spatial downsampling and lossless encoding, 0.31 bpp is obtained with 29.2 dB PSNR and 0.83 SSIM. The sensor consumes 14.8 μW (full resolution) and 4.3 μW (downsampling) at 15 fps, achieving state-of-the-art FoMs of 17.8 and 5.2 pJ/pixel∙frame respectively. Including the encoder, the overall system dissipates as low as 1.2 μJ/frame, making it an attractive solution for wireless sensor networks.",
keywords = "Bandwidth, CMOS image sensor, focal-plane compression, Image coding, Image quality, Image reconstruction, Image sensors, low bit depth, microshift, planned sensor distortion (PSD), SAR ADC, Wireless sensor networks, wireless sensor networks.",
author = "Xiaopeng Zhong and Bo Zhang and Amine Bermak and Tsui, {Chi Ying} and Law, {Man Kay}",
year = "2018",
month = "7",
day = "23",
doi = "10.1109/TCSII.2018.2859047",
language = "English",
journal = "IEEE Transactions on Circuits and Systems II: Express Briefs",
issn = "1549-7747",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - A Low-Power Compression-Based CMOS Image Sensor With Microshift-Guided SAR ADC

AU - Zhong, Xiaopeng

AU - Zhang, Bo

AU - Bermak, Amine

AU - Tsui, Chi Ying

AU - Law, Man Kay

PY - 2018/7/23

Y1 - 2018/7/23

N2 - This brief presents a low-power compression-based CMOS image sensor for wireless vision applications. The sensor implements low-bit-depth imaging with planned sensor distortion (PSD) to effectively compress both data bandwidth and processing power while maintaining high reconstruction image quality. Accordingly, a column-parallel microshift-guided successive-approximation-register (SAR) ADC is proposed to enable 3-bit PSD imaging based on a 3×3 pattern. To support normal imaging with low area overhead, the circuit is reconfigurable as an 8-bit SAR/Single-Slope (SS) ADC. The data bandwidth is further compressed by a customized lossless encoder based on predictive coding and run length coding. A 256×216 prototype imaging system composed of the compression-based image sensor and the lossless encoder is fabricated in a 0.18 μm standard CMOS process. Measurement results show that the image sensor achieves 3 bit/pixel (bpp) with 34.9 dB reconstruction peak signal-to-noise ratio (PSNR) and 0.91 structural similarity index (SSIM). With 1/4 spatial downsampling and lossless encoding, 0.31 bpp is obtained with 29.2 dB PSNR and 0.83 SSIM. The sensor consumes 14.8 μW (full resolution) and 4.3 μW (downsampling) at 15 fps, achieving state-of-the-art FoMs of 17.8 and 5.2 pJ/pixel∙frame respectively. Including the encoder, the overall system dissipates as low as 1.2 μJ/frame, making it an attractive solution for wireless sensor networks.

AB - This brief presents a low-power compression-based CMOS image sensor for wireless vision applications. The sensor implements low-bit-depth imaging with planned sensor distortion (PSD) to effectively compress both data bandwidth and processing power while maintaining high reconstruction image quality. Accordingly, a column-parallel microshift-guided successive-approximation-register (SAR) ADC is proposed to enable 3-bit PSD imaging based on a 3×3 pattern. To support normal imaging with low area overhead, the circuit is reconfigurable as an 8-bit SAR/Single-Slope (SS) ADC. The data bandwidth is further compressed by a customized lossless encoder based on predictive coding and run length coding. A 256×216 prototype imaging system composed of the compression-based image sensor and the lossless encoder is fabricated in a 0.18 μm standard CMOS process. Measurement results show that the image sensor achieves 3 bit/pixel (bpp) with 34.9 dB reconstruction peak signal-to-noise ratio (PSNR) and 0.91 structural similarity index (SSIM). With 1/4 spatial downsampling and lossless encoding, 0.31 bpp is obtained with 29.2 dB PSNR and 0.83 SSIM. The sensor consumes 14.8 μW (full resolution) and 4.3 μW (downsampling) at 15 fps, achieving state-of-the-art FoMs of 17.8 and 5.2 pJ/pixel∙frame respectively. Including the encoder, the overall system dissipates as low as 1.2 μJ/frame, making it an attractive solution for wireless sensor networks.

KW - Bandwidth

KW - CMOS image sensor

KW - focal-plane compression

KW - Image coding

KW - Image quality

KW - Image reconstruction

KW - Image sensors

KW - low bit depth

KW - microshift

KW - planned sensor distortion (PSD)

KW - SAR ADC

KW - Wireless sensor networks

KW - wireless sensor networks.

UR - http://www.scopus.com/inward/record.url?scp=85050638014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050638014&partnerID=8YFLogxK

U2 - 10.1109/TCSII.2018.2859047

DO - 10.1109/TCSII.2018.2859047

M3 - Article

AN - SCOPUS:85050638014

JO - IEEE Transactions on Circuits and Systems II: Express Briefs

JF - IEEE Transactions on Circuits and Systems II: Express Briefs

SN - 1549-7747

ER -