A Transfer Learning Framework for Anomaly Detection Using Model of Normality

Abstract

Convolutional Neural Network (CNN) techniques have proven to be very useful in image-based anomaly detection applications. CNN can be used as deep features extractor where other anomaly detection techniques are applied on these features. For this scenario, using transfer learning is common since pretrained models provide deep feature representations that are useful for anomaly detection tasks. Consequentially, anomaly can be detected by applying similarly measure between extracted features and a defined model of normality. A key factor in such approaches is the decision threshold used for detecting anomaly. While most of the proposed methods focus on the approach itself, slight attention has been paid to address decision threshold settings. In this paper, we tackle this problem and propose a welldefined method to set the working-point decision threshold that improves detection accuracy. We introduce a transfer learning framework for anomaly detection based on similarity measure with a Model of Normality (MoN) and show that with the proposed threshold settings, a significant performance improvement can be achieved. Moreover, the framework has low complexity with relaxed computational requirements.

Distance-Based Anomaly Detection for Industrial Surfaces Using Triplet Networks

“…The resulted maximum and mean distances are then subjected to an SVM with a linear kernel, with the intention of maximizing the separation between the defective test images and the non-defective test images. The maximum and mean Euclidean distances were selected based on recent work which addressed the setting of workingpoint decision thresholds for an anomaly detection task [23]. An SVM with a linear kernel allows the robust classification algorithm to be fast and memory efficient, as it is a single inner product with a O(d) prediction complexity, where d is the number of input dimensions.…”

Transfer Learning as an Essential Tool for Digital Twins in Renewable Energy Systems

“…For wind power forecasting application, we need to identify which among the X s are available in X t and which of them are relevant for T t . For anomaly detection in solar power inverter data, the first objective is to establish the MoN using anomaly free data samples [12]. So, we need to pick the anomaly free samples from the source domain data X s , Y s , to train a TL model.…”

A Critical Study on the Recent Deep Learning Based Semi-Supervised Video Anomaly Detection Methods

“…Nazare et al [14] studies the quality of features, extracted by pre-trained CNNs, for anomaly detection tasks. Ionescu et al [15] and Aburakhia et al [16] is a good example for the application of pre-trained CNNs for extracting appearance features to detect anomalies in videos. 4: Convolutional neural networks (CNNs) are suitable for processing an input data that has an inherent grid-like topology.…”