Han Xu, Zhao Xiaoli, Zeng Tao, Yang Yahan, Yu Hongyan, Zhang Cong, Wang Bin, Liu Xiaoqian, Zhang Tao, Sun Jing, Li Xinyuan, Zhao Tuo, Zhang Mingxin, Ni Yanping, Tong Yanhong, Tang Qingxin, Liu Yichun
Center for Advanced Optoelectronic Functional Materials Research, and Key Lab of UV-Emitting Materials and Technology of Ministry of Education, Northeast Normal University, 5268 Renmin Street, Changchun, 130024, China.
Department of Materials Science and Engineering, National University of Singapore, Singapore, 117575, Singapore.
Small. 2023 Feb;19(8):e2206181. doi: 10.1002/smll.202206181. Epub 2022 Dec 11.
Inspired by human eyes, the neuromorphic visual system employs a highly efficient imaging and recognition process, which offers tremendous advantages in image acquisition, data pre-processing, and dynamic storage. However, it is still an enormous challenge to simultaneously simulate the structure, function, and environmental adaptive behavior of the human eye based on one device. Here, a multimodal-synergistic-modulation neuromorphic imaging system based on ultraflexible synaptic transistors is successfully presented and firstly simulates the dry eye imaging behavior at the device level. Moreover, important functions of the human visual system in relation to optoelectronic synaptic plasticity, image erasure and enhancement, real-time preprocessing, and dynamic storage are simulated by versatile devices. This work not only simplifies the complexity of traditional neuromorphic visual systems, but also plays a positive role in the publicity of biomedical eye care.
受人类眼睛的启发，神经形态视觉系统采用了高效的成像和识别过程，这在图像采集、数据预处理和动态存储方面具有巨大优势。然而，基于单一设备同时模拟人眼的结构、功能和环境自适应行为仍然是一个巨大的挑战。在此，成功提出了一种基于超柔性突触晶体管的多模态协同调制神经形态成像系统，并首次在器件层面模拟了干眼症成像行为。此外，通过多功能器件模拟了人类视觉系统在光电突触可塑性、图像擦除与增强、实时预处理和动态存储方面的重要功能。这项工作不仅简化了传统神经形态视觉系统的复杂性，而且在生物医学眼部护理的宣传方面也发挥了积极作用。
