End-to-End Deep One-Class Learning for Anomaly Detection in UAV Video Stream - Archive ouverte HAL Access content directly
Journal Articles Journal of Imaging Year : 2021

End-to-End Deep One-Class Learning for Anomaly Detection in UAV Video Stream

(1, 2) , (2) , (2) , (3) , (4)
1
2
3
4

Abstract

In recent years, the use of drones for surveillance tasks has been on the rise worldwide. However, in the context of anomaly detection, only normal events are available for the learning process. Therefore, the implementation of a generative learning method in an unsupervised mode to solve this problem becomes fundamental. In this context, we propose a new end-to-end architecture capable of generating optical flow images from original UAV images and extracting compact spatio-temporal characteristics for anomaly detection purposes. It is designed with a custom loss function as a sum of three terms, the reconstruction loss (Rl), the generation loss (Gl) and the compactness loss (Cl) to ensure an efficient classification of the “deep-one” class. In addition, we propose to minimize the effect of UAV motion in video processing by applying background subtraction on optical flow images. We tested our method on very complex datasets called the mini-drone video dataset, and obtained results surpassing existing techniques’ performances with an AUC of 85.3.

Dates and versions

hal-03320669 , version 1 (16-08-2021)

Identifiers

Cite

Slim Hamdi, Samir Bouindour, Hichem Snoussi, Tian Wang, Mohamed Abid. End-to-End Deep One-Class Learning for Anomaly Detection in UAV Video Stream. Journal of Imaging, 2021, 7 (5), pp.90. ⟨10.3390/jimaging7050090⟩. ⟨hal-03320669⟩
18 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More