Introducing Explainable AI into an Anomaly detection for Automated Visual Inspection

Abstract

Manufacturing companies usually employ numerous inspectors for anomaly detection and it takes a high cost including time cost. Accurate and automatic anomaly detection reduces inspection cost and improves product reliability. However, these methods do not provide anomaly information. Therefore, it is difficult to relearn AI and prevent defective products from occurring. The proposed architecture explains the details of the anomaly in language and presents peripheral information for estimating the cause of the anomaly at the same time. As a result, we introduce an explainable AI for visual inspection that will enable early countermeasures. This paper reports on the results of the evaluation of a practical model for the architecture.