Abstract:

In order to improve the precision of metal magnetic memory (MMM) gradient tensor data, a method that combines the morphological filter and empirical mode decomposition(EMD) is proposed. To preserve the inter relation information between different magnetic field components while removing the noises, the data of magnetic field component is transformed into magnetic total-field strength and direction firstly. Then the data of magnetic field strength and direction is processed by the morphological filter to eliminate the influence of the instantaneous strong pulse. Finally, the processed data is decomposed with EMD to eliminate the influence of stochastic noise. An experiment on the actually acquired data is carried out to verify the proposed method, and the result shows that this method can effectively eliminate the noises in magnetic field components measurement signals and receive highprecision data of magnetic memory gradient tensor.