Using Diffractive Networks to Unlock the Power of Orbital Angular Momentum for Classification Tasks
Recent advances in optical technology have enabled the encoding of information onto light beams, such as orbital angular momentum (OAM). OAM is a quantum number that describes the angular momentum of an object in space, such as a photon. Now, researchers have demonstrated that OAM-encoded diffractive networks can be used to classify objects with high accuracy.
In a new study, researchers from the National Institute of Information and Communications Technology (NICT) in Japan used OAM-encoded diffractive networks to classify objects with significantly higher accuracy than traditional deep neural networks. The results of the study suggest that OAM-encoded diffractive networks may be a promising approach for advanced classification tasks, such as medical diagnosis and surveillance.
source: Phys.org