Constructing task visibility intervals for a surveillance system

2005

T he importance of video surveillance techniques , , has increased considerably since the latest terrorist incidents. Safety and security have become critical in many public areas, and there is a specific need to enable human operators to remotely monitor activity across large environments such as: a) transport systems (railway transportation, airports, urban and motorway road networks, and maritime transportation), b) banks, shopping malls, car parks, and public buildings, c) industrial environments, and d) government establishments (military bases, prisons, strategic infrastructures, radar centers, and hospitals).

2008

A comprehensive framework for controlling a set of Pan Tilt Zoom (PTZ) cameras is presented. The goal is to acquire close-up imagery of people in a surveillance site. The control algorithm is driven by the output of a multicamera, multi-target tracking system that performs person detection and tracking from a set of fixed surveillance cameras. This paper addresses the problem of optimally scheduling the PTZ cameras in real-time under a variety of performance objectives. The scheduler attempts to choose an optimal plan that maximizes the probability of successfully completing a biometrics task. We present a novel objective function that balances the number of captures per target and the quality of captures. We show the results of a system operating in real-time under real-world conditions on four PTZ and four fixed CCTV cameras, all controlled via a single workstation.

Lecture Notes in Computer Science, 2007

Camera networks are increasingly being deployed for security. In most of these camera networks, video sequences are captured, transmitted and archived continuously from all cameras, creating enormous stress on available transmission bandwidth, storage space and computing facilities. We describe an intelligent control system for scheduling Pan-Tilt-Zoom cameras to capture video only when task-specific requirements can be satisfied. These videos are collected in real time during predicted temporal "windows of opportunity". We present a scalable algorithm that constructs schedules in which multiple tasks can possibly be satisfied simultaneously by a given camera. We describe two scheduling algorithms: a greedy algorithm and another based on Dynamic Programming (DP). We analyze their approximation factors and present simulations that show that the DP method is advantageous for large camera networks in terms of task coverage. Results from a prototype real time active camera system however reveal that the greedy algorithm performs faster than the DP algorithm, making it more suitable for a real time system. The prototype system, built using existing low-level vision algorithms, also illustrates the applicability of our algorithms.

Autonomous Robots, 2010

We report an autonomous surveillance system with multiple pan-tilt-zoom (PTZ) cameras assisted by a fixed wide-angle camera. The wide-angle camera provides large but low resolution coverage and detects and tracks all moving objects in the scene. Based on the output of the wide-angle camera, the system generates spatiotemporal observation requests for each moving object, which are candidates for close-up views using PTZ cameras. Due to the fact that there are usually much more objects than the number of PTZ cameras, the system first assigns a subset of the requests/objects to each PTZ camera. The PTZ cameras then select the parameter settings that best satisfy the assigned competing requests to provide high resolution views of the moving objects. We propose an approximation algorithm to solve the request assignment and the camera parameter selection problems in real time. The effectiveness of the proposed system is validated in both simulation and physical experiment. In comparison with an existing work using simulation, it shows that in heavy traffic scenarios, our algorithm increases the number of observed objects by over 210%.

Machine Vision and Applications, 2007

Long gone are the days of a video surveillance system capable of processing only one video stream acquired by a single fixed camera. In those days algorithms were tested in laboratory environments, with a small number of people moving orderly and with limited clutter in the scene. Modern monitoring systems have much more demanding requirements: large, busy and complex scenes, the use of heterogeneous sensor networks, the real-time acquisition and interpretation of the evolving scene; instantaneous flagging of potentially critical situations in any weather and illumination conditions. Moreover, operators expect the real-time description of scene evolution in natural language of any type of expected and unexpected event, involving a variety of situations, from nobody in the scene to groups of people, and in some cases very crowded environments.

This paper deals with the camera control problem in automated video surveillance. We develop a solution that seeks to optimize the overall subject recognition probability by controlling the pan, tilt, and zoom of various deployed Pan/Tilt/Zoom (PTZ) cameras. Since the number of subjects is usually much larger than the number of video cameras, the problem to be addressed is how to assign subjects to these cameras. This control of cameras is based on the direction of the subject's movement and its location, distances from the cameras, occlusion, overall recognition probability so far, and the expected time to leave the site, as well as the movements of cameras and their capabilities and limitations. The developed solution works with realistic 3D environments and not just 2D scenes. We analyze the effectiveness of the proposed solution through extensive simulation.

2004

We describe the design of a scalable and wide coverage visual surveillance system. Scalability (the ability to easily add and remove cameras during system operation with minimal overhead and system degradation) is achieved by utilizing only image-based information for camera control. We show that when a Pan-Tilt-Zoom camera pans and tilts, a given image point moves in circular and linear trajectory respectively. We create a scene model using a plan view of the scene. The scene model allows us to easily handle occlusion prediction and schedule video acquisition tasks subject to visibility constraints. We describe a maximum weight matching algorithm to assign cameras to tasks that meet the visibility constraints. The system is illustrated both through simulations and real video from a 6-camera configuration.

Image Analysis and Recognition, 2006

In this work we present a new approach to learn, detect and predict unusual and abnormal behaviors of people, groups and vehicles in real-time. The proposed OBSERVER video surveillance system acquires images from a stationary color video camera and applies state-of-the-art algorithms to segment and track moving objects. The segmentation is based in a background subtraction algorithm with cast shadows, highlights and ghost's detection and removal.

2008 International Workshop on Content-Based Multimedia Indexing, 2008

In this paper, we propose novel methods for background modeling, occlusion handling and event recognition by using multi-camera configurations. Homography-related positions are utilized to construct a mixture of multivariate Gaussians to generate a background model for each pixel of the reference camera. Occlusion handling is achieved by generation of the top-view via trifocal tensors, as a result of matching over-segmented regions instead of pixels. The resulting graph is segmented into objects after determining the minimum spanning tree of this graph. Tracking of multiview data is obtained by utilizing measurements across the views in case of occlusions. Finally, the resulting trajectories are classified by GM-HMMs, yielding better results for using all different trajectories of the same object together.

2011 IEEE International Conference on Systems, Man, and Cybernetics, 2011

Surveillance is an essential part of any security operation, with cameras playing an important role. Often these cameras are controlled by a single individual, but this can lead to inefficient surveillance that leaves areas uncovered and misses targets.