Anomaly Detection with Neural Network Using a Generator

This paper concerns with the problem of detecting anomalies on X-ray images taken by Full Body Scanners (FBS). Our previous work describes the sequence of image preprocessing methods used to convert the original images, which are produced with FBS, to an images with visually distinguishable anomalies. In this paper we focus on development of the proposed methods, including the addition of preprocessing methods and the creation of generator which can produce synthetic anomalies. Examples of processed images are given. The results of using a neural network for anomaly detection are shown. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.