The invention discloses a
magnetic field imaging method with an embedded depth denoising network, and the method comprises the implementation steps: (1) building an experiment platform, arranging a main imaging
sensor array and an auxiliary sensing
sensor array in a preset two-dimensional measurement region, and constructing a
magnetic field imaging
system; (2) establishing a two-dimensional relative position mapping model between the main imaging sensor and the measured
magnetic field source based on the two-dimensional grid parameters and the
layout of the main imaging sensor; (3) under the condition that only environment
noise source interference exists,
magnetic noise signals of the main imaging sensor and the auxiliary sensing sensor are synchronously collected, and a deep denoising
learning data set is constructed; (4) constructing a deep denoising network taking a multichannel auxiliary
perception sensor
signal as an input and a main imaging sensor
noise signal as an output, and learning a mapping relation through
supervised training; (5) when a target magnetic field exists, reconstructing and suppressing a
noise component in a main imaging sensor
signal by using the trained network; and (6) mapping the denoised main imaging sensor signal into two-dimensional magnetic field distribution according to a two-dimensional relative position mapping model to realize
magnetic field imaging. According to the invention, denoising and two-dimensional imaging of magnetic field signals in a complex interference environment are realized through
deep learning noise modeling and two-dimensional
space mapping based on cooperation of multiple sensors.