Emulation of biphasic plasticity in retinal electrical synapses for light-adaptive pattern pre-processing

Digital image processing realized by memristor-based technologies

Today performance and operational efficiency of computer systems on digital image processing are exacerbated owing to the increased complexity of image processing. It is also difficult for image processors based on complementary metal-oxide-semiconductor (CMOS) transistors to continuously increase the integration density, causing by their underlying physical restriction and economic costs. However, such obstacles can be eliminated by non-volatile resistive memory technologies (known as memristors), arising from their compacted area, speed, power consumption high efficiency, and in-memory computing capability. This review begins with presenting the image processing methods based on pure algorithm and conventional CMOS-based digital image processing strategies. Subsequently, current issues faced by digital image processing and the strategies adopted for overcoming these issues, are discussed. The state-of-the-art memristor technologies and their challenges in digital image processing applications are also introduced, such as memristor-based image compression, memristor-based edge and line detections, and voice and image recognition using memristors. This review finally envisages the prospects for successful implementation of memristor devices in digital image processing.

Plasmonic Optoelectronic Memristor Enabling Fully Light-Modulated Synaptic Plasticity for Neuromorphic Vision

Exploration of optoelectronic memristors with the capability to combine sensing and processing functions is required to promote development of efficient neuromorphic vision. In this work, the authors develop a plasmonic optoelectronic memristor that relies on the effects of localized surface plasmon resonance (LSPR) and optical excitation in an Ag-TiO2 nanocomposite film. Fully light-induced synaptic plasticity (e.g., potentiation and depression) under visible and ultraviolet light stimulations is demonstrated, which enables the functional combination of visual sensing and low-level image pre-processing (including contrast enhancement and noise reduction) in a single device. Furthermore, the light-gated and electrically-driven synaptic plasticity can be performed in the same device, in which the spike-timing-dependent plasticity (STDP) learning functions can be reversibly modulated by visible and ultraviolet light illuminations. Thereby, the high-level image processing function, i.e., image recognition, can also be performed in this memristor, whose recognition rate and accuracy are obviously enhanced as a result of image pre-processing and light-gated STDP enhancement. Experimental analysis shows that the memristive switching mechanism under optical stimulation can be attributed to the oxidation/reduction of Ag nanoparticles due to the effects of LSPR and optical excitation. The authors' work proposes a new type of plasmonic optoelectronic memristor with fully light-modulated capability that may promote the future development of efficient neuromorphic vision.

A Modified SiO2-Based Memristor with Reliable Switching and Multifunctional Synaptic Behaviors

Dielectric SiO₂ has possible uses as an active layer for emerging memory due to its high on/off ratio and low operation voltage. However, SiO₂-based memory that relies on the conducting filament still has limited endurance and stability. Here, we have constructed a passivated layer of SiO₂ using Ag-doped SrTiO₃, which serves as a Ag ion reservoir for the control of filament formation. It is demonstrated that the modified memristor presents an excellent endurance switching and could stably be operated in an ambient environment for 20 days without visible degradation. Based on the reliable switching, the synaptic functions such as excitatory postsynaptic current, paired-pulse facilitation, transition from short-term memory to long-term memory, and potentiation/depression have also been implemented. Furthermore, a 7 × 7 pixel array made from memristors has successfully mimicked simple learning and forgetting behavior. The experimental results offer an alternative approach for SiO₂-based memristors and a possibility to be applied in neuromorphic computing.