Anomaly Detection in Image Datasets Using Convolutional Neural Networks, Center Loss, and Mahalanobis Distance
Garnik Vareldzhan, Kirill Yurkov, Konstantin S. Ushenin · 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology (USBEREIT) · 2021
User activities generate a significant number of poor-quality or irrelevant images and data vectors that cannot be processed in the main data processing pipeline or included in the training dataset. Such samples can be found through manual analysis by an expert or with anomalous detection algorithms. There are several formal definitions for anomalous samples. For neural networks, anomalies are usually defined as out-of-distribution samples. This work proposes methods for supervised and semi-supervised detection of out-of-distribution samples in image datasets. Our approach extends a basic neural network that solves the image classification problem. Thus, after extension one neural network can solve image classification and anomalous detection problems simultaneously. The proposed methods are based on the center loss and its effect on deep feature distribution in a last hidden layer of the neural network. This paper provides an analysis of the proposed methods for the LeNet and EfficientNet-B0 on the MNIST and ImageNet-30 datasets.