Robust Detection of Abandoned and Removed Objects in Complex Surveillance Videos

Abstract

Abstract-Tracking-based approaches for abandoned object detection often become unreliable in complex surveillance videos due to occlusions, lighting changes, and other factors. We present a new framework to robustly and efficiently detect abandoned and removed objects based on background subtraction and foreground analysis with complement of tracking to reduce false positives. In our system, the background is modeled by three Gaussian mixtures. In order to handle complex situations, several improvements are implemented for shadow removal, quick lighting change adaptation, fragment reduction, and keeping a stable update rate for video streams with different frame rates. Then, the same Gaussian mixture models used for background subtraction are employed to detect static foreground regions without extra computation cost. Furthermore, the types of the static regions (abandoned or removed) are determined by using a method that exploits context information about the foreground masks, which significantly outperforms previous edge-based techniques. Based on the type of the static regions and userdefined parameters (e.g., object size and abandoned time), a matching method is proposed to detect abandoned and removed objects. A person-detection process is also integrated to distinguish static objects from stationary people. The robustness and efficiency of the proposed method is tested on IBM Smart Surveillance Solutions for public safety applications in big cities and evaluated by several public databases such as i-Lids and PETS2006 datasets. The test and evaluation demonstrate our method is efficient to run in real-time while being robust to quick lighting changes and occlusions in complex environments.

Pixel-based colour contrast for abandoned and stolen object discrimination in video surveillance

“…Introduction: In the video surveillance domain, the automatic detection of abandoned and stolen objects in real-time has recently become a topic of great interest especially in crowded environments [1]. In general, this detection is achieved by developing a system with the following analysis stages: foreground segmentation, stationary region detection, blob classification and abandoned/stolen discrimination [1].…”

A semantic-based probabilistic approach for real-time video event recognition

“…For comparison purposes, we have selected the widely used CSM approach [3][23] [31][22] [32]. It de nes rules for each component of the event description that provide a binary decision on whether these components happened or not.…”

“…As proposed by [15], we overcome existing event recognition problems by including strategies for their solving. For the Abandoned-object PN, the left side de nes the typical model for detecting abandoned objects [31] whilst the right side describes their detection considering that the action owner is not likely to be identi ed (di cult in crowded scenarios). In addition, PN loops correspond to two temporal relations: before (before(e4, e5) and before(NOT(e5), e4)).…”

Abandoned Object Detection and Tracking in Video Surveillance

“…The attributes are used as input to an alert prioritization method which performs a ranking using alert importance. YingLiTian et al [5] proposed a new framework to robustly and efficiently detect the abandoned and removed objects in complex environments for real-time video surveillance. The mixture of Gaussians background subtraction method is employed to detect both background and static foregrounds by using the same Gaussian mixture model.…”