Automated Metal Cleanliness Analyzer (AMCA): Improving Digital Image Analysis of PoDFA Micrographs by Combining Deterministic Image Segmentation and Unsupervised Machine Learning

Quality controlQuality control of aluminumAluminum is critical for a wide range of applications across different industries. The main methodMethod for assessing aluminumAluminum cleanlinessCleanliness is PoDFAPoDFA. The manual nature of the methodMethod imposes limitations in speedSpeed and statistical robustness that made aluminumAluminum producers and suppliers call for alternative methodsMethod with higher degrees of standardization and automation in recent years. We previously demonstrated the AutomatedAutomated Metal CleanlinessMetal cleanliness Analyzer (AMCA) methodMethod as a feasible way of assessing metal cleanlinessMetal cleanliness from PoDFAPoDFA micrographs using digital deterministic image segmentationImage segmentation techniques. Here, we continue this work by combining the deterministic approach with unsupervised machine learningMachine Learning (ML) for decreasing false-positive detections and achieving a higher degree of automation. Our results show that this approach generates metal cleanlinessMetal cleanliness data closer to PoDFAPoDFA reference data than previous implementations on the one hand and decreases algorithm setup time for new types of micrographs (e.g., alloysAlloys) by automating parts of the algorithm.