A Novel Neural Network-based Data Observation Method for Zeeman Detection

Zeeman effect characterizes the phenomenon that a spectral line is split into several components in the presence of a static magnetic field. It is very important in applications, such as magnetic resonance imaging, nuclear magnetic resonance spectroscopy and atomic absorption spectroscopy. Zeeman measurement was used in Di Li's recent Nature paper to reveal a coherent magnetic field from the atomic cold neutral medium to the molecular envelope. This paper proposes a novel data observation based on neural network for Zeeman detection, which provides a different way to learn data feature automatically. Experimental results show that our method can fit the data in Li's paper well.