Vector Magnetic Field Imaging With High-Resolution TMR Sensor Arrays for Metal Structure Inspection

This paper proposes an eddy current testing (ECT) probe with high-resolution tunneling magnetoresistance (TMR) array sensors for vector magnetic field measurement. The probe consists of two printed circuit boards (PCBs), which are placed perpendicular to each other. Three arrays of TMR sensors are wire bonded on the two PCBs to measure the three components of the magnetic field. Each array contains 32 sensors and the distance between the centers of two adjacent sensors in an array is 0.5 mm. The 32 output signals of each array are multiplexed, amplified and filtered by using a circuit on the PCB. An experimental system is set up and aluminum samples with machined defects are imaged by the probe. It is seen that the patterns of the images along the x-, y- and z-axis are different, meaning the information in the three images can compensate each other to a certain extent. The images change their orientations if the defects are oriented along different angles. An image processing algorithm including image segmentation, normalization, edge detection and K-means classification is developed to class the defects. It is seen that the defects can be classified accurately with the algorithm. Then, the orientations of the defects are calculated from the images with a linear regression of the angles of the three axis images. It is found that the mean error of the orientation quantification is only 4.14°, which is much smaller than the mean error calculated from each axis image separately.