Researchers develop system to improve latent fingerprint recognition
Recently, a research team led by Professor Long Shibing from University of Science and Technology of China (USTC) of the Chinese Academy of Sciences, in collaboration with Professor Liu Qi from Fudan University, developed developed an in-sensor reservoir computing system for latent fingerprint recognition with deep-ultraviolet imaging synapses and a memristor array. This study was published in natural communication.
Deep ultraviolet (DUV) photodetectors play a pivotal role in deep space exploration, environment controland biological information identification. However, conventional field DUV fingerprint recognition systems use separate sensors, memory, and processors, which significantly increases latency in the field. decision and thus overall computing power.
Inspired by the human visual perception system, the research team built an RC system in the DUV sensor with optical synapses as the input layer of the reservoir and a memristor device array as the reading network, can sense and process in parallel to ensure high efficiency and low power consumption.
The team used a Ga-rich component design and developed amorphous GaOx (a-GaOx) imaging synapses with enhanced photoconductive continuum (PPC) effects. The non-linear mapping relationship for the RC system in the DUV sensor is built by inputting 4-bit equivalent light pulses for the simulation so that the image pixel sequence information can be sampled for calculated values. power.
Finally, training of the reservoir outputs was achieved through the stable polymorphic modulation properties of the memristor device array, enabling small-scale DUV fingerprint recognition. The excellent recognition accuracy of the DUV fingerprint image when using this dual feature strategy and hardware system is close to the simulation results.
The system achieves 100% recognition accuracy after 100 training epochs and maintains 90% accuracy even with 15% ambient noiseconsistent with the noise-cancelling properties of DUV lamps.
This hardware-packed DUV in-sensor RC system provides a good reference prototype for efficient identification and secure applications of latent fingerprints. It is also an important reference for the development of intelligent optoelectronic devices in the DUV range.
“This prototype system… will provide more insight into emerging sensor-in-reservoir reservoir computing. Overall, the subject matter of this work is really interesting,” said one reviewer. natural communication.
Zhongfang Zhang et al., Sensor-Contained Computing System for Latent Fingerprint Recognition with Deep Ultraviolet Image Synapses and Memristor Arrays, natural communication (2022). DOI: 10.1038/s41467-022-34230-8
Provided by University of Science and Technology of China
quote: Researchers develop system to improve latent fingerprint recognition (2022, December 2) retrieved December 2, 2022 from https://techxplore.com/news/2022 -12-latent-fingerprint-recognition.html
This document is the subject for the collection of authors. Other than any fair dealing for private learning or research purposes, no part may be reproduced without written permission. The content provided is for informational purposes only.