One-Class Classification in Images and Videos Using a Convolutional Autoencoder With Compact Embedding

Abstract

In One-Class Classification (OCC) problems, the classifier is trained with samples of a class considered normal, such that exceptional patterns can be identified as anomalies. Indeed, for real-world problems, the representation of the normal class in the feature space is an important issue, considering that one or more clusters can describe different aspects of the normality. For classification purposes, it is important that these clusters be as compact (dense) as possible, for better discriminating anomalous patterns, which is a recurrent problem in OCC tasks. This work introduces a hybrid approach using deep learning and One-Class Support Vector Machine (OC-SVM) methods, named Convolutional Autoencoder with Compact Embedding (CAE-CE), for enhancing the compactness of clusters in the feature space. Such an approach is still underexplored in the literature, being restricted to models within the context of metric learning. Additionally, the absence of anomalous samples during training makes it difficult to determine when to interrupt the learning process, so as to avoid over-compression of the normal examples, thus resulting in overfitting of the model. In this work, we propose a novel sensitivity-based stop criterion, and its suitability for OCC problems was assessed. Using an OC-SVM for the classification task, several experiments were done using publicly available image and video datasets. We also introduce other two new benchmarks, specifically designed for video anomaly detection in highways. The final performance of the proposed method was compared with a baseline Convolutional Autoencoder (CAE). Overall results suggest that the enhanced compactness introduced by the CAE-CE improved the classification performance for most datasets. Also, the qualitative analysis of frames at the visual level indicated that features learned by CAE-CE are closely correlated to the anomalous events.

Deep-Compact-Clustering Based Anomaly Detection Applied to Electromechanical Industrial Systems

“…Although the DAE model has the ability to generate a feature space mapping at the output of the bottleneck coding layer, DAEs may be inefficient when applied directly to OCC problems, since the feature space mapping of the bottleneck can be sparse, i.e., it does not guarantee a compact mapping of the data in the bottleneck, which is an essential issue in OCC problems for a desirable result [ 36 ]. In this regard, it is applied as a framework for enhancing the compactness of clusters in the feature space.…”

“…Further-more, there is a wide margin of separation between normal samples and anomalous samples and therefore outside the classification boundary. This effect on performance improvement is also addressed in [ 36 ].…”

Intrusion Detection Systems for IoT: opportunities and challenges offered by Edge Computing and Machine Learning

“…Anomaly detection belongs to a family of machine learning tasks named one-class classification, in which a ML model is trained to perform a binary prediction on whether or not an input sample belongs to a particular class, and Auto Encoders are widely used for this goal [119] [120] [121]. The usual anomaly detection strategy to follow when using an AE, is to train the model only on non-anomalous samples.…”

Low-Power Detection and Classification for In-Sensor Predictive Maintenance Based on Vibration Monitoring

“…Moreover, if the input data are underlying normal operations of the monitored apparatus, anomalous inputs can be detected by comparing them with the corresponding reconstructed outputs. However, the limitation of the AE is the capability to recognize what it shall be normal, namely, AEs are only able to detect if an anomaly occurs, but they are not capable to classify it [7]. As it will be discussed in the next section, several solutions have been presented in the related works to overcome the above limitation by recurring to conventional and more compact Machine Learning (ML) approaches or to more complex DL, which usually are challenged in finding an acceptable trade-off between the number of physical resources needed by their implementations, the processing speed, and the detection/classification accuracy.…”