Bio-Inspired Ferroelectric Adaptive Transistors for Intelligent Vision Systems

To address the challenges of low data processing efficiency, spatiotemporal information separation, and high energy consumption in traditional machine vision systems, this work proposes a bio-inspired ferroelectric adaptive transistor based on annealing-free HZO ferroelectric films. The FeTFT achieves the lowest fabrication temperature while demonstrating exceptional performance with a high ON/OFF ratio of ${3}.{8}\times {10} ^{{9}}$ and a large memory window of 2.86 V. The FeTFT exhibits bio-synaptic optoelectronic co-response characteristics and implements biological adaptive functions through dynamic reconfiguration of ferroelectric polarization. By constructing a fire vision system based on the spatiotemporal fusion mechanism, the FeTFT achieves 100% motion direction recognition accuracy and three-level speed classification capability. This research establishes a novel paradigm for developing low-power, dynamically adaptable bio-inspired intelligent vision systems.