Abandoned Object Detection in Video-Surveillance: Survey and Comparison

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.

Spatiotemporal event detection: a review

“…Object-based event extraction methods are based on frame differencing, optical flow, and background subtraction (Parekh, Thakore, and Jaliya 2014; Wang and Snoussi 2014; Yu and Yang 2017;Luna et al 2018), and the extracted objects can be further classified or tracked. For example, Yu and Yang (2017) identified dust storm features by developing a region-growing approach and integrating heuristic and machine learning-based multi-thresholds of dust concentration.…”

Robust Real-time Detection of Abandoned Objects using a Dual Background Model

“…There have been many studies for the detection of abandoned objects using a single background subtraction model. One of the disadvantages is that it is difficult to select and track a stationary object among all the foregrounds generated from a single background model [1,2].…”

“…In particular, There is a growing interest in automatic abandonment detection such as suspicious abandoned object detection, garbage dumping detection, and illegal parking detection, etc. Although there have been many studies on the detection of abandoned objects, there are still several challenging issues such as occlusions, illumination changes, owner's re-attendance, and long-term abandonment [1,2]. In recent years, implementing video analysis functions on edge devices, rather than on high-performance servers, has become an important challenge [3,4].…”

“…They extract a candidate stationary object from the model in their own way and then track the state of the object for a predetermined period analyaing consecutive foreground information generated from the model. If the tracking is successful, they determine the object as an abandoned object [2]. However, due to the nature of the background subtraction technique, there is a problem that the foreground is gradually absorbed into the background.…”

Object Detection Techniques in Videos

“…In [11], The work explored on deep learning technique reputation which is built on CNN for object detection methods. Different techniques for object detection systems are also compared and assessed here.…”