Defect Classification in Additive Manufacturing Using CNN-Based Vision Processing
This work addresses quality control in additive manufacturing, but it is incremental as it applies existing CNN and active learning methods to a new domain-specific dataset.
The paper tackles defect classification in additive manufacturing by applying CNNs to image datasets and incorporating active learning to reduce training data requirements, achieving accurate classification with a human-in-the-loop mechanism.
The development of computer vision and in-situ monitoring using visual sensors allows the collection of large datasets from the additive manufacturing (AM) process. Such datasets could be used with machine learning techniques to improve the quality of AM. This paper examines two scenarios: first, using convolutional neural networks (CNNs) to accurately classify defects in an image dataset from AM and second, applying active learning techniques to the developed classification model. This allows the construction of a human-in-the-loop mechanism to reduce the size of the data required to train and generate training data.