Digitization of Biomimetic Vision Sensor Based on the Common Housefly (Musca Domestica)

摘要

In comparison to traditional camera-eye visual system of mammals such as human beings, the compound vision systemof a common house fly, Musca domestica, is capable of enhanced motion detection and tracking. Most computer visionsystems today have higher spatial resolution but lower motion detection and tracking capabilities when compared tocompound vision. In applications requiring obstacle avoidance and quick visual data processing, compound vision ismore suitable as compared to normal mammalian inspired vision systems. Proof-of-concept has shown that even withoutusing a computer processing system, compound vision sensors can mimic the motion hyperacuity characteristic of thefly’s visual system and at the same time provide near instantaneous edge detection and motion tracking results. Whilethese early prototypes were successfully implemented in device-level analog components, we were interested indetermining if a similar system could be implemented as an embedded system. The first step of this process isdetermining the best way to digitize the signals without losing the key feature of compound vision, motion hyperacuity.