Researchers from UCLA, Electrical and Computer engineers have developed a physical artificial neural network that was modeled the state the device let how to work on the human brain. So that can analyze large volumes of data and detects the objects at the actual speed of light and this device was created using a 3D printer at the UCLA Samueli School of Engineering.
Nowadays all the latest devices in everyday life use computerized cameras to identify objects as that of automated teller machines that can detect handwritten values like that of cheque deposits or internet search engines that instantly match photos to other similar images in their servers.
The device developed by UCLA called “diffractive deep neural network,” which uses the light bouncing from the object itself to identify that object in as little time as it would take a little more time to identify things.
The UCLA device does not need advanced computing programs to process an image of the object and to identify to which category object belongs by undergoing a process through optical sensors and no energy is consumed to run the device because it only uses diffraction of light. This technology based on the invention that is also used in microscopic imaging and medicine. Aydogan Ozcan, the study’s principal investigator and the UCLA Chancellor’s Professor of Electrical and Computer Engineering
Aydogan Ozcan, Professor of Electrical and Computer Engineering at UCLA stated that Artificial Intelligence implemented in this device analyzes data, images and classify objects and this optical artificial neural network device is particularly designed to know the functionality of the brain which can be further developed to enable new camera designs and unique optical components that work
Ozcan also said, “The researchers then implemented a network using a computer to detect the objects in front of it by installing the pattern of diffracted light which is designed exactly as a glass maze”. By the experiments, the researchers also demonstrated that the device could accurately identify handwritten numbers and items of clothing. As its components are designed using a 3D printer, the artificial neural network device with hundreds of millions of artificial neurons identify objects where components can be made inexpensively.