UAV-AdNet: Unsupervised Anomaly Detection using Deep Neural Networks for Aerial Surveillance
This work addresses the problem of detecting unusual observations in critical infrastructure surveillance using UAVs, representing an incremental advancement by integrating GPS and image data.
The paper tackles anomaly detection in aerial surveillance by proposing UAV-AdNet, a deep neural network that combines GPS and image data for unsupervised anomaly detection, showing improved performance in scene reconstruction and anomaly detection tasks compared to baselines.
Anomaly detection is a key goal of autonomous surveillance systems that should be able to alert unusual observations. In this paper, we propose a holistic anomaly detection system using deep neural networks for surveillance of critical infrastructures (e.g., airports, harbors, warehouses) using an unmanned aerial vehicle (UAV). First, we present a heuristic method for the explicit representation of spatial layouts of objects in bird-view images. Then, we propose a deep neural network architecture for unsupervised anomaly detection (UAV-AdNet), which is trained on environment representations and GPS labels of bird-view images jointly. Unlike studies in the literature, we combine GPS and image data to predict abnormal observations. We evaluate our model against several baselines on our aerial surveillance dataset and show that it performs better in scene reconstruction and several anomaly detection tasks. The codes, trained models, dataset, and video will be available at https://bozcani.github.io/uavadnet.