High-Endurance STO:YSZ Optoelectronic Memristors with Vertically Aligned Nanocomposite Structure for Edge Detection.

Edge detection plays a critical role in cutting-edge domains such as real-time monitoring and automatic driving, with optoelectronic device-based real-time image processing garnering significant attention. However, the poor endurance and unstable optical responsivity of conventional optoelectronic memristors constrain their application in highly integrated edge detection systems. In this study, a silicon-based integrated optoelectronic memristor based on SrTiO3:(Y2O3:ZrO2) (STO:YSZ) vertically aligned nanocomposite (VAN) structure is introduced, where the conductive channels at the spatial vertical interface providing an effective transport pathway. The device achieves excellent endurance (108 switching cycles) and stable multi-band (405-650 nm) optical switching. Additionally, it also exhibits the ability of simulating biological synaptic plasticity, implementing optical image extraction and ASCII code transmission. Importantly, edge detection of real-time road vehicle imagery is demonstrated via an optoelectronic memristor network. This work opens a promising paradigm for developing stable and high endurance machine vision systems based on optoelectronic memristor.