1.School of Electronic and Information Engineering,Ningbo University of Technology, Ningbo 315211，China
2.Ningbo Institute of Materials Technology and Engineering,Chinese Academy of Sciences,Ningbo 315201，China
3.Center for Excellence in Brain Science and Intelligence Technology, Chinese Academy of Sciences,Shanghai 200031, China
4.College of Materials Science and OptoElectronic Technology, University of Chinese Academy of Sciences,Beijing 100049, China
et al.
Conventional artificial visual systems are based on the von Neumann architecture and CMOS technologies, in which image perception, memory and processing units are physically separated, thus facing great challenges of realizing high-density integration, low power consumption and high speed.Inspired by biological visual systems, neuromorphic visual systems based on optoelectronic memristors can realize integrated sensing-computing-memory as well as parallel processing of visual information, which becomes one of the most important developing directions of artificial visual systems.They can not only mimic retinal functions such as responding directly to light and performing the preprocessing of visual information like image contrast enhancement and noise reduction, but also mimic the functions of the visual cortex, i.e., visual perception including pattern recognition and classification. This review focuses on the research progress of optoelectronic memristors and their applications in artificial visual systems. First, the features of memristor are introduced. Second, the research progress of optoelectronic memristors is reviewed in terms of materials, structures, working mechanisms and synaptic function emulation. Then, the applications of optoelectronic memristors in artificial visual systems are discussed. Finally, the problems of optoelectronic memristors which need be resolved for practical applications are summarized.