Anomaly Detection in Unsupervised Surveillance Setting Using Ensemble of Multimodal Data with Adversarial Defense

TL;DR

This paper presents an unsupervised ensemble approach combining image and sensor data for anomaly detection in drone surveillance, incorporating adversarial defense to enhance robustness and security.

Contribution

It introduces a novel ensemble method using CNN regression and autoencoders for multimodal anomaly detection with adversarial defense in autonomous drone surveillance.

Findings

Achieved 97.8% accuracy on IEEE SP Cup-2020 dataset.

Validated robustness on an in-house dataset.

Demonstrated effectiveness of multimodal ensemble in real-time anomaly detection.

Abstract

Autonomous aerial surveillance using drone feed is an interesting and challenging research domain. To ensure safety from intruders and potential objects posing threats to the zone being protected, it is crucial to be able to distinguish between normal and abnormal states in real-time. Additionally, we also need to consider any device malfunction. However, the inherent uncertainty embedded within the type and level of abnormality makes supervised techniques less suitable since the adversary may present a unique anomaly for intrusion. As a result, an unsupervised method for anomaly detection is preferable taking the unpredictable nature of attacks into account. Again in our case, the autonomous drone provides heterogeneous data streams consisting of images and other analog or digital sensor data, all of which can play a role in anomaly detection if they are ensembled synergistically. To that end, an ensemble detection mechanism is proposed here which estimates the degree of abnormality of analyzing the real-time image and IMU (Inertial Measurement Unit) sensor data in an unsupervised manner. First, we have implemented a Convolutional Neural Network (CNN) regression block, named AngleNet to estimate the angle between a reference image and current test image, which provides us with a measure of the anomaly of the device. Moreover, the IMU data are used in autoencoders to predict abnormality. Finally, the results from these two pipelines are ensembled to estimate the final degree of abnormality. Furthermore, we have applied adversarial attack to test the robustness and security of the proposed approach and integrated defense mechanism. The proposed method performs satisfactorily on the IEEE SP Cup-2020 dataset with an accuracy of 97.8%. Additionally, we have also tested this approach on an in-house dataset to validate its robustness.