An insect eye-based image sensor with very large field of view

In this paper we discuss the design of a novel miniaturized image sensor based on the working principle of insect facet eyes. The main goals are to design an imaging system which captures a large field of view (FOV) and to find a good trade-off between image resolution and sensitivity. To capture a total FOV of 124°, we split up this FOV into 25 different zones. Each of these angular zones is imaged by an isolated optical channel on our image sensor. There is an overlap between the zones to cover the full FOV but the different zones are imaged on separated regions at the image sensor. Every optical channel in the designed component consists of two lenses that are tilted with respect to each other and the optical axis. Because of this tilt of the lenses, we are able to minimize field curvature and distortion in the obtained images at the detector, and have an angular resolution below 1°. The optical system was implemented and optimized in the ray-tracing program ASAP. The parameters (in one channel) that are optimized to obtain this large FOV with a good image resolution and sensitivity are the radius of curvature of the two lenses, their conical factor and their tilt in two directions with respect to the optical axis of the complete system. The lenses are each placed on a pedestal that connects the lens to a planar substrate. We also add absorbing tubes that connect the two lenses in one channel to eliminate stray-light between different optical channels. The obtained image quality of the design is analyzed using our simulation model. This is determined by different parameters as there are: modulation transfer function, distortion, sensitivity, angular resolution, energy distribution in each channel and channel overlap. The modulation transfer function shows us that maximum contrast in the image is reached up to 0.3LP/°, distortion is maximal 21% in one of the 25 different channels, the sensitivity is 0.3% and the resolution is better than 1°.