Ultra-Thin Image Sensor Being Developed Based on Insect Eye

The principle of hyperacuity has now been effectively incorporated in a technical model by German scientists.

Insects have inspired scientists to adapt characteristics that have been optimized over millions of years to present-day devices. Research scientists from the Fraunhofer Institute for Applied Optics and Precision Engineering IOF (Jena, Germany), for example, are working on the development of an ultra-thin image sensor based on the insect eye. Dr. Andreas Brückner improved the imaging properties of these systems with regard to sensor applications.

Insects have not just two, but thousands of eyes. Each facet of their eye picks up one image point, and the many facets, each with its own lens and visual cells, are spread over the surface of a hemisphere. As a result, the insect eye can cover a wide viewing angle--but the resolution of the images produced is not particularly high. This is unexpected, given that insects can fly very precise maneuvers. They are able to do so because of the principle of hyperacuity--insects see more than the images captured by their compound eyes because the visual fields of adjacent facets overlap, and Dr. Brückner is replicating this phenomenon in a technical system.

"The aim was to develop micro-optical compound eyes which contain numerous parallel imaging channels and which are also extremely compact, thinner than 0.5 millimeters,” reported Dr. Brückner. To accomplish this, he began by studying how images are generated in artificial compound eyes. Given that each facet captures one image point, the challenge was to accomplish controlled overlapping in the technical system. With a precise knowledge of the angular sensitivity, image signals of adjacent facets can then be compared with each other. This makes it possible to determine the position of the object viewed in a two-dimensional visual field with an accuracy that is many times higher than the image resolution.

A comparison has shown that an artificial compound eye lens can transfer information with an effective image resolution of 625 x 625 pixels, although the number of actually available image pixels is limited to 50 x 50. As a result, the sensor can recognize simple objects, precisely determine their position and size, and reliably detect movements.

Several projects are already underway to implement the process, for instance, as solar altitude sensors in automobiles, for recognizing traffic lanes in driver assistance systems, and in machine vision.


Related Links:
Fraunhofer Institute for Applied Optics and Precision Engineering IOF