Minimizing Soil Mineralization Effects in Pulse Induced Metal Detector Through Dual Adaptive Filter-Based Compensation to Detect Low Metal Contents

De-mining operations are crucial for humanitarian efforts and safety in conflict-affected regions. As conflicts persist, so do advancements in technology and the challenges they present, particularly with landmines containing very low or sub-gram metal contents. In this paper, we address the challenge of improving the accuracy of mine detection systems in mineralized soil. We present a novel algorithm based on dual adaptive filters to compensate the effect of soil mineralization. The filters learn the response of the mineralization content within the ground through supervised learning and remove the effects with a high degree of confidence (above 98%). The paper also considers a challenging mineralized environment containing the baked clay bricks and shows the effectiveness of the presented dual adaptive filter-based compensation (DAFC) algorithm for it. The parametric approach used in the algorithm also provides a way to reduce the effect of over compensation in real-time.