Researchers in the Department of Electronic Systems Engineering at Tokyo University of Sciences (TUS) has developed an industrial self -operating synapses that mimic nerve clamps in the brain, able to distinguish between colors with a resolution of up to 10 nm, which is the accuracy of the human eye capacity.
The device derives its self -energy through solar cells without the need for external energy sources, which makes it suitable for merged into low -energy peripherals, such as smartphones, wearable devices, and self -driving vehicles.
The researchers relied on the design of the device to integrate two types of solar cells sensitive to dyes, responding to different wavelengths of light, allowing the implementation of logical functions based on the contrast of wavelengths.
In a practical experiment, the researchers used the device within the physical reservoir computing system, as it was able to identify human motor patterns, such as walking or raising the hand, using light signals in red, green and blue colors, and the device achieved a resolution of 82% in distinguishing between 18 different patterns of movements and colors.
Dr. Takshi Acono, a member of the research team, said the results show promising potential for this innovation in enhancing the capabilities of low -energy visual artificial intelligence systems.
