AI Camera to Mimic Human Brain
Artificial Intelligence would soon have a mind of its own, being able to process images and mimic the human retina according to an article published in the Scientific American journal in June 2017.
AI today requires large sophisticated computers, which have the potential packed into a small electronic device. Researchers in Europe are developing a camera with brain-like algorithms and light sensors, to process images and mimic the human retina, respectively. Just as much as an AI camera would account for an attractive smartphone feature, it would improve the efficacy of self-driving and autonomous flying drones.
The new camera is expected to be more cost-effective and power-efficient than the conventional digital cameras and is called an ultralow-power event-based camera, or ULPEC, as it is made up of pixel sensors that come to life only to record a new image. Conventional cameras waste memory and battery life by capturing repetitive data which does not change frame to frame.
This new AI camera will also devises new electrical components allowing it to react to any change in light or movement within microseconds. This reaction occurs only when the light striking the pixel sensors exceeds the present threshold.
The photo sensors would be made up of tiny semiconductors and circuitry on silicon, which converts the light into electrical signals and sends it to the neural network.
Cohesive circuits and a new electronic component – memristor, would act as the equivalent of synaptic connections in the brain. The memristors will consist of a thin layer of ferroelectric material which will be placed between two electrodes.
The resistance of memristors can be tuned using voltage, which will help the neural network learn, just as the brain learns on stimulation of synapses. The amalgamation of bio-inspired optical sensors and neural networks would make the camera suitable for self-driving cars and autonomous drones,
“In self-driving cars the onboard computer must react to changes very quickly while navigating through traffic or determining the movement of pedestrians,” Posch explains. “The ULPEC can detect and process these changes rapidly.” German automotive equipment manufacturer Bosch—also involved in the project—will investigate how the camera might be used as part of its autonomous and computer-aided driving technology.
You may be interested
New Study Links Plastic Degradation to Emission of Greenhouse GasesMohit Joshi - December 12, 2018
Researchers from University of Hawaii at Manoa revealed that degradation of plastic leads to emission of several greenhouse gases. Plastic is a widely used material, owing to…
Hearing Aid Market Segmented by Type, Market Demand Industry Growth, Trends and Forecasts 2018 to 2023Albert R - December 11, 2018
Worldwide Market Reports recently released "Hearing Aid Market Report 2017 to 2023". Most recent market Value and Sales Price in the Hearing Aid Market report classifies the…
Hair drier 2018 – 2023 Future Demand Analysis, Industry sizes, competition, Market overviewsAlbert R - December 11, 2018
Worldwide Market Reports recently released "Hair drier Market Report 2017 to 2023". Most recent market Value and Sales Price in the Hair drier Market report classifies the…