Adaptive Hodgkin–Huxley Neuron for Retina‐Inspired Perception

Abstract

Biomimetic machine vision systems require structures, including an artificial retina to capture and process images, and an artificial network that imitates the cortex for visual inspection and identification tasks. One of the key challenges for biomimetic perceptional machine visual systems is the self‐adaptive behavior of the retina, which is known as the ability to adjust and eventually perceive objects through a wide range of environmental light intensities. To achieve this, current systems use either complex processing algorithms or specific devices with optoelectronic properties which results in complexity and less efficiency. Herein, an artificial retina circuit is reported to imitate the adaptation process to light (photopic and scotopic adaptation) of the perceptional biovisual system. The implementation of an artificial Hodgkin–Huxley neuron circuit based on adaptive VO2 memristors has helped the artificial retina to successfully demonstrate time‐varying activation and inhibition characteristics. The adaptable perception function under different illumination conditions is faithfully achieved by the artificial HH neuron based on the self‐adaptability of the memristor. The function is consistent with the dynamic light amplitude ranges of biological vision systems. More importantly, this development shows a pervasive way to design and fabricate a universal environmentally adaptable artificial system.

Nanowire-based synaptic devices for neuromorphic computing

“…By adjusting the operation parameters, resistive switching properties between high resistance state and low resistance state (LRS) are realized. Volatile and nonvolatile memristors have been widely studied, where the volatile memristors usually have a short retention time on LRS and the nonvolatile memristors usually have a longer one [40]. Ni/NiO core-shell NWs were fabricated into crossbar memristor [36].…”

Tunable Anti-Ambipolar Vertical Bilayer Organic Electrochemical Transistor enable Neuromorphic Retinal Pathway

“…They also cannot compete with the biological realism of Hodgkin Huxley(HH) neuron models, which closely replicates the ow of sodium and potassium ions within a biological neuron, serving as the fundamental mechanisms to generate action potentials 12 . Such biologically realistic circuit models have been demonstrated in-silico for pattern classi cation, arti cial retinas, and neurological disease models 13,14 . The sodium channel ion ux exhibits an inherent anti-ambipolar function which is critical for generating an action potential and must be replicated in neuromorphic systems 10,12 .…”

Ga2O3 Bipolar Heterojunction-Based Optoelectronic Synapse Array with Visual Attention

“…Recently, artificial visual neurons (AVNs) endowed with the essential feature of synaptic plasticity were achieved in lateral field-effect transistors, vertical memristor devices, or photodetectors. For instance, intelligent tasks, such as handling complex image processing like visual adaptation, − visual pattern recognition, − and collision avoidance, were mimicked by these devices for enhancing recognition accuracy. , Especially, a hardware-level photoelectronic reservoir computing (RC) system, combining photosynapse and a non-volatile memristor array, has also been innovatively developed by Zhang et al to recognize the latent fingerprint with in-sensor and parallel in-memory computing . Recently, the introducing attention mechanism has been demonstrated in an optoelectronic synapse transistor based on a ReS 2 /hBN/monolayer graphene heterojunction, which allows for selective accentuation of relevant information that enhances the capability in multi-target recognition .…”