GAN based ball screw drive picture database enlargement for failure classification
This work provides a method for manufacturing engineers to overcome data scarcity for failure detection, which is a common bottleneck for applying machine learning in industrial settings.
This paper addresses the scarcity of failure image datasets in manufacturing by generating synthetic images of ball screw surface failures (pitting and rust) using a Generative Adversarial Network (GAN). The generated images were evaluated for quality and diversity using expert observation, t-SNE, and FID scores, and were shown to positively impact classification test performance when used to augment or replace real image datasets.
The lack of reliable large datasets is one of the biggest difficulties of using modern machine learning methods in the field of failure detection in the manufacturing industry. In order to develop the function of failure classification for ball screw surface, sufficient image data of surface failures is necessary. When training a neural network model based on a small dataset, the trained model may lack the generalization ability and may perform poorly in practice. The main goal of this paper is to generate synthetic images based on the generative adversarial network (GAN) to enlarge the image dataset of ball screw surface failures. Pitting failure and rust failure are two possible failure types on ball screw surface chosen in this paper to represent the surface failure classes. The quality and diversity of generated images are evaluated afterwards using qualitative methods including expert observation, t-SNE visualization and the quantitative method of FID score. To verify whether the GAN based generated images can increase failure classification performance, the real image dataset was augmented and replaced by GAN based generated images to do the classification task. The authors successfully created GAN based images of ball screw surface failures which showed positive effect on classification test performance.