Machine learning based localization and classification with atomic magnetometers

Abstract

We demonstrate identification of position, material, orientation and shape of objects imaged by an ${}^{85}$Rb atomic magnetometer performing electromagnetic induction imaging supported by machine learning. Machine learning maximizes the information extracted from the images created by the magnetometer, demonstrating the use of hidden data. Localization 2.6 times better than the spatial resolution of the imaging system and successful classification up to 97$\%$ are obtained. This circumvents the need of solving the inverse problem, and demonstrates the extension of machine learning to diffusive systems such as low-frequency electrodynamics in media. Automated collection of task-relevant information from quantum-based electromagnetic imaging will have a relevant impact from biomedicine to security.

Main text: 5 pages, 4 figures. Supplementary material: 7 pages, 8 figures. Minor typos amended and References re-arrangement. Added details on algorithms performance in Supplemental Material. Results unchanged. Published version available at https://doi.org/10.1103/PhysRevLett.120.033204

References in corpus (10)

• Finding Density Functionals with Machine Learning
• Identifying structural flow defects in disordered solids using machine learning methods
• Quantum Loop Topography for Machine Learning
• Entanglement-Based Machine Learning on a Quantum Computer
• Electromagnetic induction imaging with a radio-frequency atomic magnetometer
• Quantum machine learning over infinite dimensions
• Identifying polymer states by machine learning
• Eddy current imaging with an atomic radio-frequency magnetometer
• Magnetic induction tomography using an all-optical ${}^{87}$Rb atomic magnetometer
• Optical atomic magnetometry for magnetic induction tomography of the heart

Cited by in corpus (10)

• Light scattering control in transmission and reflection with neural networks
• Automatic spin-chain learning to explore the quantum speed limit
• Active Underwater Detection with an Array of Atomic Magnetometers
• Sub-picotesla widely tunable atomic magnetometer operating at room-temperature in unshielded environments
• Sub-Sm${}^{-1}$ electromagnetic induction imaging with an unshielded atomic magnetometer
• Electromagnetic induction imaging with atomic magnetometers: unlocking the low-conductivity regime
• Self-adaptive loop for external disturbance reduction in differential measurement set-up
• Detection and characterisation of conductive objects using electromagnetic induction and a fluxgate magnetometer
• Burst Eddy Current Testing with a Diamond Magnetometry
• Identification of object composition with Magnetic Inductive Tomography