Learned Anomaly Detection with Terahertz Radiation in Inline Process Monitoring

Abstract Terahertz tomographic imaging as well machine learning tasks represent two emerging fields in the area of nondestructive testing. Detecting outliers measurements that are caused by defects is main challenge inline process monitoring. An efficient control enables to intervene directly during manufacturing and, consequently, reduce product discard. We focus on plastics and ceramics, for which terahertz radiation perfectly suited because its characteristics, propose a density based technique automatically detect anomalies measured data. The algorithm relies classification method learning. For verification, supervised data generated measuring system approximates an process. experimental results show use radiation, combined with algorithm, has great potential real In further investigation additional simulated enlarge set, especially variety defects. model propagation means Eikonal equation.