Optimal human labelling for anomaly detection in industrial inspection

Anomaly detection with machine learning in industrial inspection systems for manufactured products relies on labelleddata. This rises the question how the labelling by humans should be conducted. We consider the case where we want to optimise the cost of the combined inspection process done by humans and
an algorithm. This also influences the combined performance of the trained model as well as the knowledge of the performance of this model. We focus on so called one-class classification problem models which produce a continuous outlier score. We establish some cost model for human and machine combined inspection of samples. We then discuss in this cost model how to select two optimal boundaries of the outlier score where in between these two boundaries human inspection takes place. We also frame this established knowledge into an applicable algorithm.