Foveated imaging systems for target-centric imaging inspired by animal eyes

Avian eyes have evolved a deep central fovea, enabling efficient refraction of light for magnified imaging and motion tracking. This adaptation is crucial for detecting and tracking objects in dynamic environments. Additionally, avian eyes can perceive a broad spectrum of light, including visible and ultraviolet wavelengths, aiding in differentiating target objects from complex backgrounds. Despite advancements in artificial vision systems inspired by animal vision, the exceptional object detection and targeting capabilities of avian eyes, utilizing foveated and multi-spectral imaging, remain underexplored. In this study, we present an artificial vision system leveraging these unique features of avian vision. Our system incorporates an artificial fovea and vertically-stacked perovskite photodetector arrays, optimized through simulations to demonstrate foveated and multi-spectral imaging. Our system successfully identifies colored and mixed-color objects and detects remote objects through foveated imaging. Furthermore, we discuss its potential applications in unmanned aerial vehicles requiring detection, tracking, and recognition of distant targets in dynamic environments.