Ultralow-Power Machine Vision with Self-Powered Sensor Reservoir.

Jie Lao, Mengge Yan, Bobo Tian, Chunli Jiang, Chunhua Luo, Zhuozhuang Xie, Qiuxiang Zhu, Zhiqiang Bao, Ni Zhong, Xiaodong Tang, Linfeng Sun, Guangjian Wu, Jianlu Wang, Hui Peng, Junhao Chu, Chungang Duan
Author Information
  1. Jie Lao: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  2. Mengge Yan: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  3. Bobo Tian: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China. ORCID
  4. Chunli Jiang: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  5. Chunhua Luo: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  6. Zhuozhuang Xie: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  7. Qiuxiang Zhu: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  8. Zhiqiang Bao: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  9. Ni Zhong: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  10. Xiaodong Tang: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  11. Linfeng Sun: Centre for Quantum Physics Key Laboratory of Advanced Optoelectronic Quantum Architecture and Measurement (MOE), School of Physics Beijing Institute of Technology, Beijing, 100081, China.
  12. Guangjian Wu: Institute of Optoelectronics, Frontier Institute of Chip and System, Fudan University, 220 Handan Road, Shanghai, 200433, China.
  13. Jianlu Wang: Institute of Optoelectronics, Frontier Institute of Chip and System, Fudan University, 220 Handan Road, Shanghai, 200433, China.
  14. Hui Peng: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  15. Junhao Chu: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.
  16. Chungang Duan: Key Laboratory of Polar Materials and Devices (MOE), Ministry of Education, Department of Electronics, East China Normal University, Shanghai, 200241, China.

Abstract

A neuromorphic visual system integrating optoelectronic synapses to perform the in-sensor computing is triggering a revolution due to the reduction of latency and energy consumption. Here it is demonstrated that the dwell time of photon-generated carriers in the space-charge region can be effectively extended by embedding a potential well on the shoulder of Schottky energy barrier. It permits the nonlinear interaction of photocurrents stimulated by spatiotemporal optical signals, which is necessary for in-sensor reservoir computing (RC). The machine vision with the sensor reservoir constituted by designed self-powered Au/P(VDF-TrFE)/Cs AgBiBr /ITO devices is competent for both static and dynamic vision tasks. It shows an accuracy of 99.97% for face classification and 100% for dynamic vehicle flow recognition. The in-sensor RC system takes advantage of near-zero energy consumption in the reservoir, resulting in decades-time lower training costs than a conventional neural network. This work paves the way for ultralow-power machine vision using photonic devices.

Keywords

References

  1. Nat Commun. 2021 Mar 19;12(1):1798 [PMID: 33741964]
  2. Nat Commun. 2016 May 04;7:11502 [PMID: 27143121]
  3. Nat Commun. 2017 Dec 19;8(1):2204 [PMID: 29259188]
  4. Nat Nanotechnol. 2019 Aug;14(8):776-782 [PMID: 31308498]
  5. Natl Sci Rev. 2020 Jul 25;8(2):nwaa172 [PMID: 34691573]
  6. Nano Lett. 2019 Mar 13;19(3):2066-2073 [PMID: 30803237]
  7. Adv Mater. 2019 May;31(22):e1900326 [PMID: 31025419]
  8. Nat Commun. 2017 May 12;8:15199 [PMID: 28497781]
  9. Adv Sci (Weinh). 2022 May;9(15):e2106092 [PMID: 35285175]
  10. Sci Adv. 2021 May 14;7(20): [PMID: 33990331]
  11. Nat Commun. 2020 Jun 25;11(1):3211 [PMID: 32587241]
  12. Nat Commun. 2021 Jan 18;12(1):408 [PMID: 33462233]
  13. Sci Adv. 2021 Mar 17;7(12): [PMID: 33731346]
  14. Nature. 2020 Jan;577(7792):641-646 [PMID: 31996818]
  15. Nature. 2020 Mar;579(7797):62-66 [PMID: 32132692]
  16. Nat Nanotechnol. 2018 May;13(5):404-410 [PMID: 29632398]
  17. ACS Nano. 2020 Jan 28;14(1):746-754 [PMID: 31887010]
  18. Sci Adv. 2020 Jun 24;6(26):eaba6173 [PMID: 32637614]
  19. Nature. 2015 May 7;521(7550):61-4 [PMID: 25951284]

Grants

  1. 62174053/NSFC
  2. 61804055/NSFC
  3. 62004204/NSFC
  4. 19JC141670/Shanghai Science and Technology Innovation Action Plan
  5. 21JC1402000/Shanghai Science and Technology Innovation Action Plan
  6. 21520714100/Shanghai Science and Technology Innovation Action Plan
  7. 2021MD0AB03/Open Research Projects of Zhejiang Lab
  8. 18ZR1410900/Natural Science Foundation of Shanghai

Word Cloud

Created with Highcharts 10.0.0reservoirvisionin-sensorenergymachinesystemcomputingconsumptionRCdevicesdynamicneuromorphicvisualintegratingoptoelectronicsynapsesperformtriggeringrevolutionduereductionlatencydemonstrateddwelltimephoton-generatedcarriersspace-chargeregioncaneffectivelyextendedembeddingpotentialwellshoulderSchottkybarrierpermitsnonlinearinteractionphotocurrentsstimulatedspatiotemporalopticalsignalsnecessarysensorconstituteddesignedself-poweredAu/PVDF-TrFE/CsAgBiBr/ITOcompetentstatictasksshowsaccuracy9997%faceclassification100%vehicleflowrecognitiontakesadvantagenear-zeroresultingdecades-timelowertrainingcostsconventionalneuralnetworkworkpaveswayultralow-powerusingphotonicUltralow-PowerMachineVisionSelf-PoweredSensorReservoirCs2AgBiBr6in-sensorslead-freedoubleperovskites

Similar Articles

Cited By (18)