The structure and mechanism of the human visual system contain rich treasures, and surprising effects can be achieved by simulating the human visual system. In this article, starting from the human visual system, we compare and discuss the discrepancies between the human visual system and traditional machine vision systems. Given the wide variety and large volume of visual information, the use of non-von Neumann structured, flexible neuromorphic vision sensors can effectively compensate for the limitations of traditional machine vision systems based on the von Neumann architecture. Firstly, this article addresses the emulation of retinal functionality and provides an overview of the principles and circuit implementation methods of non-von Neumann computing architectures. Secondly, in terms of mimicking the retinal surface structure, this article introduces the fabrication approach for flexible sensor arrays. Finally, this article analyzes the challenges currently faced by non-von Neumann flexible neuromorphic vision sensors and offers a perspective on their future development.

人类视觉系统的结构和机制蕴藏着丰富的宝藏，通过模拟人类视觉系统可以达到令人惊奇的效果。本文从人类视觉系统出发，比较和讨论人类视觉系统与传统机器视觉系统的差异。鉴于视觉信息种类繁多、数量庞大，使用非冯·诺依曼结构、灵活的神经拟态视觉传感器可以有效弥补基于冯·诺依曼架构的传统机器视觉系统的局限性。首先，本文讨论了视网膜功能的仿真，并概述了非冯诺依曼计算架构的原理和电路实现方法。其次，在模拟视网膜表面结构方面，介绍了柔性传感器阵列的制作方法。最后，本文分析了非冯诺依曼柔性神经形态视觉传感器目前面临的挑战，并对其未来发展提出了展望。