Automated segmentation and analysis of microscopy images of laser powder bed fusion melt tracks

With the increasing adoption of metal additive manufacturing (AM), researchers and practitioners are turning to data-driven approaches to optimise printing conditions. Cross-sectional images of melt tracks provide valuable information for tuning process parameters, developing parameter scaling data, and identifying defects. Here we present an image segmentation neural network that automatically identifies and measures melt track dimensions from a cross-section image. We use a U-Net architecture to train on a data set of 62 pre-labelled images obtained from different labs, machines, and materials coupled with image augmentation. After tuning the neural network hyperparameters, the learned model has a pixelwise accuracy of over 99% and an F1 score over 90%. The neural network exhibits robustness when tested on images captured by various users, printed on different machines, and acquired using different microscopes. A post-processing module extracts the height and width of the melt pool, and the wetting angles. We discuss opportunities to improve model performance and avenues for transfer learning, such as extension to other AM processes such as directed energy deposition. • Trained a model to segment optical micrographs of LPBF melt track cross-sections. • The model has an accuracy greater than 99% and F1 score greater than 90%. • Associated software converts the segmented images into quantitative metrics. • The model and software are available as an installable Python package. • The model can be used for transfer learning to apply it to other AM techniques.