Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island

In-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence (AI) technology, enabling intelligent vision sensing and extensive data processing. These architectures must demonstrate potential advantages in te...

Full description

Saved in:
Bibliographic Details
Main Authors: Kyungmoon Kwak, Kyungho Park, Jae Seong Han, Byung Ha Kang, Dong Hyun Choi, Kunho Moon, Seok Min Hong, Gwan In Kim, Ju Hyun Lee, Hyun Jae Kim
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:International Journal of Extreme Manufacturing
Subjects:
Online Access:https://doi.org/10.1088/2631-7990/adebbe
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849720049737138176
author Kyungmoon Kwak
Kyungho Park
Jae Seong Han
Byung Ha Kang
Dong Hyun Choi
Kunho Moon
Seok Min Hong
Gwan In Kim
Ju Hyun Lee
Hyun Jae Kim
author_facet Kyungmoon Kwak
Kyungho Park
Jae Seong Han
Byung Ha Kang
Dong Hyun Choi
Kunho Moon
Seok Min Hong
Gwan In Kim
Ju Hyun Lee
Hyun Jae Kim
author_sort Kyungmoon Kwak
collection DOAJ
description In-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence (AI) technology, enabling intelligent vision sensing and extensive data processing. These architectures must demonstrate potential advantages in terms of mass production and complementary metal oxide semiconductor compatibility. Here, we introduce a visible-light-driven neuromorphic vision system that integrates front-end retinomorphic photosensors with a back-end artificial neural network (ANN), employing a single neuro-inspired indium-gallium-zinc-oxide phototransistor (NIP) featuring an aluminum sensitization layer (ASL). By methodically adjusting the ASL coverage on IGZO phototransistors, a fast-switching response-type and a synaptic response-type of IGZO phototransistors are successfully developed. Notably, the fabricated NIP shows a remarkable retina-like photoinduced synaptic plasticity under wavelengths up to 635 nm, with over 256-states, weight update nonlinearity below 0.1, and a dynamic range of 64.01. Owing to this technology, a 6 × 6 neuro-inspired optical image sensor array with the NIP can perform highly integrated sensing, memory, and preprocessing functions, including contrast enhancement, and handwritten digit image recognition. The demonstrated prototype highlights the potential for efficient hardware implementations in in-sensor AI technologies.
format Article
id doaj-art-e058f4fcf1fe4fc297da27d56ce93f00
institution DOAJ
issn 2631-7990
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series International Journal of Extreme Manufacturing
spelling doaj-art-e058f4fcf1fe4fc297da27d56ce93f002025-08-20T03:12:02ZengIOP PublishingInternational Journal of Extreme Manufacturing2631-79902025-01-017606550410.1088/2631-7990/adebbeSwiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization islandKyungmoon Kwak0Kyungho Park1Jae Seong Han2Byung Ha Kang3Dong Hyun Choi4Kunho Moon5Seok Min Hong6Gwan In Kim7Ju Hyun Lee8Hyun Jae Kim9https://orcid.org/0000-0002-6879-9256School of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaDepartment of Chemical Engineering, Massachusetts Institute of Technology , Cambridge, Massachusetts 02139, United States of AmericaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaDepartment of Electrical and Computer Engineering, University of Illinois Urbana-Champaign , Urbana, Illinois 61801, United States of AmericaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaSchool of Electrical and Electronic Engineering, Yonsei University , 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of KoreaIn-optical-sensor computing architectures based on neuro-inspired optical sensor arrays have become key milestones for in-sensor artificial intelligence (AI) technology, enabling intelligent vision sensing and extensive data processing. These architectures must demonstrate potential advantages in terms of mass production and complementary metal oxide semiconductor compatibility. Here, we introduce a visible-light-driven neuromorphic vision system that integrates front-end retinomorphic photosensors with a back-end artificial neural network (ANN), employing a single neuro-inspired indium-gallium-zinc-oxide phototransistor (NIP) featuring an aluminum sensitization layer (ASL). By methodically adjusting the ASL coverage on IGZO phototransistors, a fast-switching response-type and a synaptic response-type of IGZO phototransistors are successfully developed. Notably, the fabricated NIP shows a remarkable retina-like photoinduced synaptic plasticity under wavelengths up to 635 nm, with over 256-states, weight update nonlinearity below 0.1, and a dynamic range of 64.01. Owing to this technology, a 6 × 6 neuro-inspired optical image sensor array with the NIP can perform highly integrated sensing, memory, and preprocessing functions, including contrast enhancement, and handwritten digit image recognition. The demonstrated prototype highlights the potential for efficient hardware implementations in in-sensor AI technologies.https://doi.org/10.1088/2631-7990/adebberetinomorphic hardwarein-sensor preprocessingimage recognitionneuro-inspired optical sensorsindium-gallium-zinc-oxidemetallic sensitization layer
spellingShingle Kyungmoon Kwak
Kyungho Park
Jae Seong Han
Byung Ha Kang
Dong Hyun Choi
Kunho Moon
Seok Min Hong
Gwan In Kim
Ju Hyun Lee
Hyun Jae Kim
Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
International Journal of Extreme Manufacturing
retinomorphic hardware
in-sensor preprocessing
image recognition
neuro-inspired optical sensors
indium-gallium-zinc-oxide
metallic sensitization layer
title Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
title_full Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
title_fullStr Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
title_full_unstemmed Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
title_short Swiftly accessible retinomorphic hardware for in-sensor image preprocessing and recognition: IGZO-based neuro-inspired optical image sensor arrays with metallic sensitization island
title_sort swiftly accessible retinomorphic hardware for in sensor image preprocessing and recognition igzo based neuro inspired optical image sensor arrays with metallic sensitization island
topic retinomorphic hardware
in-sensor preprocessing
image recognition
neuro-inspired optical sensors
indium-gallium-zinc-oxide
metallic sensitization layer
url https://doi.org/10.1088/2631-7990/adebbe
work_keys_str_mv AT kyungmoonkwak swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT kyunghopark swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT jaeseonghan swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT byunghakang swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT donghyunchoi swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT kunhomoon swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT seokminhong swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT gwaninkim swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT juhyunlee swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland
AT hyunjaekim swiftlyaccessibleretinomorphichardwareforinsensorimagepreprocessingandrecognitionigzobasedneuroinspiredopticalimagesensorarrayswithmetallicsensitizationisland