An Optoelectronically Modulated Artificial Bionic Visual Neuron

The visual perception system is the most important part of the human body to perceive and receive information about the external environment, in which the visual neuron is the basic unit of the visual perception system to achieve the function. As an important part of the bionic neuromimetic perception system, the artificial bionic visual neural system is diligently striving to surmount the computational bottlenecks inherent in the traditional von Neumann architecture. However, structural redundancy and information transmission loss remain a challenge for artificial visual perception systems based on sensor-artificial neuron connections. Here, we propose an artificial bionic visual neuron of sense-storage-computing monolithic type, integrating a light-absorbing layer based on chalcogenide quantum dots and conjugated polymers on top of tantalum oxide artificial neural components, which has good optoelectronic synergistic integration capability. The proposed artificial visual neurons exhibit typical bipolar switching characteristics and integral ignition performance, while showing threshold change behaviour for photomodulation. The unique optoelectronic co-integration ability of the artificial visual neurons can provide assistance for constructing more efficient and low-power artificial visual perception systems, which has a broad application prospect.