Design and testing of a multi-aperture bio-inspired sensor

摘要

The goal of this research is to develop an imaging system based on the visual system of the common housefly, Musca domestica. While traditional imaging systems (which resemble the human visual system) are well suited for a wide variety of image processing tasks, sensors based on the fly's multi-aperture visual system show promise in specific applications, including motion and edge detection. This thesis quantifies, in signal processing terms, how the fly is able to achieve motion acuity much higher than its photoreceptor spacing suggests (a phenomenon called "hyperacuity"). Next, this information is used to optimize motion hyperacuity in an optical model of a new, more compact fly eye sensor. The optical model is used to design a hardware implementation of the sensor. Finally, the newly designed sensor is shown to outperform fly eye sensors previously developed at the University of Wyoming.