Patchwise Sparse Dictionary Learning from pre-trained Neural Network Activation Maps for Anomaly Detection in Images

In this work, we investigate a methodology to perform anomaly detection and localization on images. The method leverages both sparse representation learning and the adoption of a pre-trained neural network for classification purposes. The objective is to assess the effectiveness of the K-SVD sparse dictionary learning algorithm and understand the role of neural network activation maps as data descriptors. We extract meaningful representation features and build a sparse dictionary of the most expressive ones. The dictionary is built only over features coming from images without anomalies. Thus, images containing anomalies will either have a non-sparse representation as linear combinations of the dictionary elements or a high reconstruction error. We show that the proposed pipeline achieves state-of-the-art performance in terms of AUC-ROC score over benchmarks such as MVTec Anomaly Detection, Rd-MVTec Anomaly Detection, Magnetic Tiles Defect, and BeanTech Anomaly Detection Datasets.