Self-Powered Synapse Brings Human-Like Vision to AI Devices – Neuroscience News
Innovative strides have been made by researchers in the realm of machine vision, marking a significant breakthrough in how artificial systems perceive color. A groundbreaking self-powered artificial synapse, capable of recognizing colors with near-human precision, has been developed. Unlike traditional systems that require substantial external energy and data processing, this new device mimics biological vision and independently generates electricity using solar cells.
The device’s remarkable capability includes distinguishing colors with a 10-nanometer resolution and performing logical functions based on light wavelengths. Such technology upends expectations, paving the way for low-power, high-performance machine vision in edge devices—including smartphones, wearables, and autonomous vehicles.
As artificial intelligence (AI) progresses, so too does the role of machine vision—a critical enabler in modern technology advancements. Despite significant progress, current machine vision systems face challenges. Processing the vast amounts of visual data generated each second demands extensive power, storage, and computational resources. These requirements make deploying visual recognition in edge devices, like smartphones and drones, challenging.
Conversely, the human visual system presents a compelling blueprint. It cleverly filters and prioritizes visual information, allowing for efficient processing with minimal power usage. Inspired by this system, neuromorphic computing—an approach that mimics biological neural structures and functions—offers a promising path to overcoming existing computer vision challenges.
Nevertheless, two primary challenges have constrained breakthroughs in machine vision. One is achieving color recognition that rivals human vision, and the other is eliminating dependency on external power sources, thus minimizing energy consumption.
In this context, a research group led by Associate Professor Takashi Ikuno from the School of Advanced Engineering, Department of Electronic Systems Engineering at Tokyo University of Science (TUS), Japan, offers a visionary solution. Their research, presented in Volume 15 of Scientific Reports in May 2025, reveals a self-powered artificial synapse capable of distinguishing colors with astonishing accuracy.
Developed with the collaboration of Mr. Hiroaki Komatsu and Ms. Norika Hosoda from TUS, the device integrates two distinct dye-sensitized solar cells that respond variably to different light wavelengths. Unlike conventional systems reliant on external power, the proposed synapse harnesses solar energy, rendering it particularly suited to edge computing applications where energy efficiency is paramount.
As demonstrated through exhaustive testing, this system can discern colors with a resolution of 10 nanometers over the visible spectrum. This level of discrimination nears the capabilities of the human eye. Furthermore, the device exhibits bipolar responses, generating positive voltage under blue light and negative voltage under red light, facilitating complex logic operations typically requiring multiple devices.
Dr. Ikuno highlights, “The results show great potential for applying this next-generation optoelectronic device, which enables high-resolution color discrimination and logical operations simultaneously, to low-power artificial intelligence (AI) systems with visual recognition.”
In practical demonstrations, the team utilized their device within a physical reservoir computing framework to recognize varying human movements recorded in red, green, and blue hues. Their system achieved an impressive 82% accuracy in classifying 18 different combinations of colors and movements, demonstrating its efficiency with just a single device compared to the multiple photodiodes required by conventional systems.
The implications of this research are vast. Autonomous vehicles could significantly benefit, enhancing their ability to identify traffic lights, road signs, and obstacles more efficiently. In healthcare, wearable devices equipped with this technology could monitor vital signs like blood oxygen levels, minimizing battery drain. In consumer electronics, smartphones, as well as augmented and virtual reality headsets, could enjoy improved battery life while maintaining sophisticated visual recognition capabilities.
Dr. Ikuno adds, “We believe this technology will contribute to realizing low-power machine vision systems with color discrimination capabilities close to those of the human eye, with applications in optical sensors for self-driving cars, low-power biometric sensors for medical use, and portable recognition devices.”
This research heralds a significant advancement in the field of computer vision, setting the stage for everyday devices to perceive the world as humans do.
Funding: The work was partially supported by the JST and the establishment of university fellowships for the creation of science and technology innovation (Grant Number JPMJFS2144). Additional support was provided by the JST SPRING (Grant Number JPMJSP2151).
Author: Yoshimasa Iwasaki
Source: Tokyo University of Science
Contact: Yoshimasa Iwasaki – Tokyo University of Science
Image: The image is credited to Neuroscience News
Original Research: Open access. “Polarity-tunable dye-sensitized optoelectronic artificial synapses for physical reservoir computing-based machine vision” by Takashi Ikuno et al. Scientific Reports.
Abstract: Conventional machine vision systems process huge time-series data per second, presenting significant challenges for edge-device applications due to limitations in data transfer and storage. Inspired by the human visual system, artificial optoelectronic synapses replicating synaptic responses have emerged as promising solutions. However, achieving color recognition comparable to human vision remains challenging. Moreover, most optoelectronic artificial synapses rely on photocurrent-based operation, producing low current values and necessitating external circuits.
This study reports a self-powered optoelectronic artificial synapse capable of distinguishing wavelengths with a resolution of 10 nm by integrating dye-sensitized solar cells. The device exhibits synaptic responses to light pulses and bipolar responses when exposed to different wavelengths. The wavelength-dependent bipolar behavior enables exceptional separation capabilities, achieving six-bit resolution with 64 distinct states and supporting multiple logic operations, including AND, OR, and XOR, within a single device. Additionally, the device leverages distinct responses to red, green, and blue light irradiation for physical reservoir computing, facilitating the classification of color-coded human motion with an accuracy of 82%.
These findings advance the development of optoelectronic artificial synapses for precise, human-eye-like color discrimination.
