Abandoned Object Detection via Temporal Consistency Modeling and Back-Tracing Verification for Visual Surveillance

Abstract

This paper presents an effective approach for detecting abandoned luggage in surveillance videos. We combine short-and long-term background models to extract foreground objects, where each pixel in an input image is classified as a 2-bit code. Subsequently, we introduce a framework to identify static foreground regions based on the temporal transition of code patterns, and to determine whether the candidate regions contain abandoned objects by analyzing the back-traced trajectories of luggage owners. The experimental results obtained based on video images from 2006 Performance Evaluation of Tracking and Surveillance and 2007 Advanced Video and Signal-based Surveillance databases show that the proposed approach is effective for detecting abandoned luggage, and that it outperforms previous methods. Index Terms-Abandoned luggage detection, abandoned object detection, short-term background model, long-term background model, object detection and tracking, visual surveillance.

Security Event Recognition for Visual Surveillance

“…As a practical application topic, security event recognition attracted much less effort from researchers. And even most of the existing works for this topic only focus on the shallow task of detecting abandoned luggage (Porikli et al, 2007, Fan and Pankanti, 2011, Liao et al, 2008, Evangelio et al, 2011, Fan et al, 2013, Lin et al, 2015. They learn a robuster background model and then identify static foreground objects by subtraction.…”

“…To handle temporary occlusion in finding the owner, (Lin et al, 2015) used a back-tracing verification strategy. However, the verification is triggered only when there is no moving foreground object within the object's neighbor region of predefined radius.…”

“…We evaluate our method on the benchmark PETS2006 2 and PETS2007 3 . A new dataset called ABODA provided by Lin et al (Lin et al, 2015) is also used for further test. The results show that our algorithm outperforms the state-of-the-art methods for abandoned object detection inference.…”

Abandoned Object Detection and Tracking in Video Surveillance

“…Kevin Lin et al [1] proposed a temporal consistency model combining a back-tracing algorithm for abandoned object detection. The temporal consistency model is described by a very simple Finite state machine (FSM).…”

Abandoned Object Detection Using Dual Background Model from Surveillance Videos

“…If hitCount goes above the user defined threshold value, an alarm is triggered indicating the abandoned object is detected. Kevin Lin, Shen-Chi Chen, Chu-Song Chen, Daw-Tung Lin and Yi-Ping Hung [18] have presented their work on abandoned object detection via temporal consistency modelling and back-tracking verification for visual surveillance. Temporal dual-rate foreground integration method is proposed for static-foreground estimation for single camera video images.…”