Machine Learning Based Localization and Classification with Atomic Magnetometers - Shape and Position

Datasets created with an atomic magnetometer performing magnetic induction imaging, and later analysed with machine learning.In this dataset, object's position is randomly varied in the (x,y) plane, and four different shapes are used.Shape 01: 50 mm x 50 mm x 3 mm Aluminium squareShape 02: 50 mm diameter x 3 mm Aluminium diskShape 03: 50 mm x 25 mm x 3 mm Aluminium rectangleShape 04: 50 mm base isosceles triangle, 50 mm height.Data are arranged per shape; position is indicated in the file's name, in mm.Four classes of data are collected: Radius (R), Phase (P), in-phase (X), and out-of-phase (Y). Blind datasets: Information removed from filename.Details are reported in the file: Shape_Position_blind_files.pdfFor further details, see the paper published in Physical Review Letters: https://doi.org/10.1103/PhysRevLett.120.033204.