KAUST researchers have developed an artificial electronic retina that can “see” and recognize handwritten numbers, an advance with potential computer vision applications.
Mani Teja Vijjapu, an electrical engineer Ph.D. student, Khaled Nabil Salama and colleagues are said to have designed and fabricated a series of photoreceptors that detect the intensity of visible light through a change in electrical capacitance, and mimic the behavior of the eye’s rod retinal cells.
When connected to an electronic CMOS sensing circuit and a spiking neural network (a single-layer network with 100 output neurons), the array was able to recognize 70 percent of handwritten numbers.
“The ultimate goal of our research in this area is to develop efficient neuromorphic vision sensors to build efficient cameras for computer vision applications,” Salama said in a statement. “Existing systems use photo detectors that require power to operate and therefore consume a lot of energy, even in standby. In contrast, our proposed photoreceptors are capacitive devices that do not consume static current for their operation.”
MORE FROM ELECTRONICS & COMMUNICATIONS
According to KAUST (King Abdullah University of Science & Technology, Saudi Arabia), the photoreceptor array is made by sandwiching a material with suitable optical and dielectric properties between a lower aluminum electrode and a patterned upper electrode of indium tin oxide to form a pixelated array of miniature photosensitive metal-insulator-metal capacitors. The array is made on a thin polyimide substrate, so it is flexible and can be bent, including into a semicircular shape that mimics the human eye.
KAUST said the team used a hybrid material of perovskite (methylammonium lead bromide (MAPbBr3)) nanocrystals embedded in terpolymer polyvinylidene fluoride trifluoroethylene-chlorofluoroethylene (PVDF-TrFE-CEF). MAPbBr3 strongly absorbs 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 wearer life and high wearer mobility,” says Vijjapu.
Tests with a 4×4 array and LED illumination of different visible colors indicate that the optical response of the array mimics the response of the human eye with maximum sensitivity to green light. Importantly, the photoreceptors also appear to be very stable, with no change in response after 129 weeks of storage in ambient conditions.
Future plans for the team include building larger arrays of photoreceptors, optimizing the interface circuit design, and using a multi-layered neural network to improve the accuracy of recognition functionality.
The team’s findings were published in Natural light: science and applications.
