A Magnetic Gradient Tensor Based Method for UXO Detection on Movable Platform
Xiaoping Zheng, Yichen Tian, Bin Wang
Abstract
A positioning method based on magnetic gradient tensor was proposed for UXO detection on movable platforms. A multi-resolution linear regression algorithm was developed to select measurement points collected when the instrument was relatively stable. In this algorithm, Pearson Correlation Coefficient (PCC) was adopted as a quality factor for characterizing the stability of instrument. Then, selected measurement points were processed in a modified Euler method for UXO position inversion. The results from the modified Euler method were selected again based on normalized magnetic source strength (NSS) and clustered by the DBSCAN algorithm. Validity of the method was tested with both simulation data and field experiment data collected by a backpack gradiometer. The test result showed that all 4 targets were detected and positioning errors were all lower than 0.15m.