Abstract
During the last few years, abandoned object detection has emerged as a hot topic in the video-surveillance community. As a consequence, a myriad of systems has been proposed for automatic monitoring of public and private places, while addressing several challenges affecting detection performance. Due to the complexity of these systems, researchers often address independently the different analysis stages such as foreground segmentation, stationary object detection, and abandonment validation. Despite the improvements achieved for each stage, the advances are rarely applied to the full pipeline, and therefore, the impact of each stage of improvement on the overall system performance has not been studied. In this paper, we formalize the framework employed by systems for abandoned object detection and provide an extensive review of state-of-the-art approaches for each stage. We also build a multi-configuration system allowing one to select a range of alternatives for each stage with the objective of determining the combination achieving the best performance. This multi-configuration is made available online to the research community. We perform an extensive evaluation by gathering a heterogeneous dataset from existing data. Such a dataset allows considering multiple and different scenarios, whereas presenting various challenges such as illumination changes, shadows, and a high density of moving objects, unlike existing literature focusing on a few sequences. The experimental results identify the most effective configurations and highlight design choices favoring robustness to errors. Moreover, we validated such an optimal configuration on additional datasets not previously considered. We conclude the paper by discussing open research challenges arising from the experimental comparison.
Highlights
Developing automated video-surveillance systems is attracting huge interests for monitoring public and private places
For the foreground segmentation stage, we have considered Mixture of Gaussian (MoG) [28] as the baseline, since it is widely used across the Abandoned Object Detection (AOD) literature, and we have included alternative approaches not employed within the context of AOD systems, such as K-nearest neighbors background subtraction (KNN) and Pixel-based Adaptive Word Consensus Segmenter (PAWCS)
We evaluate the performance for each stage of the AOD system over all sequences in Table 5 where the abandoned objects lifespan is at least 30 s, i.e., 18 events
Summary
Developing automated video-surveillance systems is attracting huge interests for monitoring public and private places. As these systems become larger, effectively observing all cameras in a timely manner becomes a challenge, especially for public and crowded places such as airports, buildings, or railway stations. Deep learning models [39,40] have recently emerged as promising frameworks to unify modeling and feature selection. These models are limited to employing training and test data from the same video sequence.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
