Artificial retina enables better computer vision

Researchers at the King Abdullah University of Science and Technology (KAUST) have fabricated an array of photoreceptors that detect the intensity of visible light via a change in electrical capacitance, mimicking the behaviour of the eye’s rod retina cells. When the array was connected to an electronic CMOS-sensing circuit and a spiking neural network (a single-layer network with 100 output neurons) it was able to recognise handwritten numbers with an accuracy of around 70%.

“The ultimate goal of our research in this area is to develop efficient neuromorphic vision sensors to build efficient cameras for computer vision applications,” said researcher Khaled Nabil Salama. “Existing systems use photodetectors that require power for their operation and thus consume a lot of energy, even on standby. In contrast, our proposed photoreceptors are capacitive devices that don’t consume static power for their operation.”

The photoreceptor array is made by sandwiching a material with suitable optical and dielectric properties between a bottom aluminium electrode and a patterned top electrode of indium tin oxide to form a pixelated array of miniature light-sensitive metal-insulator-metal capacitors. The array is made on a thin substrate of polyimide so that it is flexible and can be curved as desired, including a hemispherical shape mimicking the human eye.

In selecting materials for their photoreceptor, the researchers used a hybrid material of perovskite (methylammonium lead bromide, or MAPbBr₃) nanocrystals embedded in terpolymer polyvinylidene fluoride trifluoroethylene-chlorofluoroethylene (PVDF-TrFE-CEF). MAPbBr₃ is a strong absorber of visible light, while PVDF-TrFE-CEF has a high dielectric constant.

“We chose hybrid perovskites because of their exceptional photoelectronic properties, such as excellent light absorption, long carrier lifetime and high carrier mobility,” said PhD student Mani Teja Vijjapu.

Tests with a 4x4 array and LED illumination of different visible colours indicate that the optical response of the array mimics the response of the human eye with a maximum sensitivity to green light. The photoreceptors were also found to be highly stable, with no change in response even after being stored for 129 weeks in ambient conditions. The results were published in the journal Light: Science & Applications.

Future plans for the team include building larger arrays of photoreceptors, optimising the interface circuit design and employing a multilayered neural network to improve the accuracy of the recognition functionality.

Image credit: ©stock.adobe.com/au/valerybrozhinsky

Originally published here.