A PoDFA Benchmarking Study Between Manual and AI-supervised Machine Learning Methods to Evaluate Inclusions in Wrought and Foundry Aluminum Alloys

The PoDFAPoDFA inclusionInclusions measurement is achieved by identifying the inclusionsInclusions and their concentration in the melt for each type with a trained operator. The standard technique is realized by using a square grid with an optical microscope to count the total area with each detected square. This manual and non-efficient methodMethod requires a lot of time and effort and can generate important variations in PoDFAPoDFA results for reproducibility and repeatability. In the past, there were many unsuccessful attempts to automatically detect, count, and classify all inclusionInclusions types due to the complexity of the application. Disc sampling, image artifacts, polishing defectsDefects, and metallurgical constituents are some examples that can interfere with the inclusionInclusions detection and the measurement methodology. Commercial image analysis systems with threshold options and Boolean logical operationsOperation are not sufficient to automate the solution. The implementation of artificial intelligenceArtificial intelligence technologies such as supervised machine learningMachine Learning (ML) algorithms are necessary to automate this complex methodMethod. The benchmarking study was achieved between the standard PoDFAPoDFA methodology compared to the artificial intelligent way. Results show that the new technique exhibits a good correlation and a high potential for industrial use.