Siamese Basis Function Network for Data Efficient Defect Classification in Technical Domains

Training Deep Learning Models in technical domains often brings the challenges that although the task is clear, insufficient data for training is available. In this work we propose a novel approach based on the combination of Siamese-Networks and Radial-Basis- Function-Networks to perform data-efficient classification without pre-Training by measuring the distance between images in semantic space in a data efficient manner. We develop the models using three technical datasets, the NEU dataset the BSD dataset as well as the TEX dataset. Additional to the technical domain show the general applicability to classical datasets (cifar10 and MNIST) as well. The approach is tested against state of the art models (Resnet50 and Resnet101) by stepwise reducing the number of samples available for training. The authors show that the proposed approach outperforms the state of the art models in the low data regime.