Henry Detmold

Measuring Latency for Video Surveillance Systems

The increased flexibility and other benefits offered by IP network cameras makes them a common choice for installation in new and expanded surveillance networks. One commonly quoted limitation of IP cameras is their high latency when compared to their analogue counterparts. This causes some reluctance to install or upgrade to digital cameras, and is slowing the adoption of live, intelligent analysis techniques in video surveillance systems. This paper presents methods for measurement of the latency in systems based upon digital IP or analogue cameras. These methods are camera-agnostic and require no specialised hardware. We use these methods to compare a variety of camera models. The results demonstrate that whilst analogue cameras do have a lower latency, most IP cameras are within acceptable tolerances. The source of the latency within an IP camera is also analysed, with prospects for improvement identified.

Finding camera overlap in large surveillance networks

Recent research on video surveillance across multiple cameras has typically focused on camera networks of the order of 10 cameras. In this paper we argue that existing systems do not scale to a network of hundreds, or thousands, of cameras. We describe the design and deployment of an algorithm called exclusion that is specifically aimed at finding correspondence between regions in cameras for large camera networks. The information recovered by exclusion can be used as the basis for other surveillance tasks such as tracking people through the network, or as an aid to human inspection. We have run this algorithm on a campus network of over 100 cameras, and report on its performance and accuracy over this network.

Scalable Surveillance Software Architecture

Video surveillance is a key technology for enhanced protection of facilities such as airports and power stations from various types of threat. Networks of thousands of IP-based cameras are now possible, but current surveillance methodologies become increasingly ineffective as the number of cameras grows. Constructing software that efficiently and reliably deals with networks of this size is a distributed information processing problem as much as it is a video interpretation challenge. This paper demonstrates a software architecture approach to the construction of large scale surveillance network software and explores the implications for instantiating surveillance algorithms at such a scale. A novel architecture for video surveillance is presented, and its efficacy demonstrated through application to an important class of surveillance algorithms.

Automatic camera placement for large scale surveillance networks

Automatic placement of surveillance cameras in arbitrary buildings is a challenging task, and also one that is essential for efficient deployment of large scale surveillance networks. Existing approaches for automatic camera placement are either limited to a small number of cameras, or constrained in terms of the building layouts to which they can be applied. This paper describes a new method for determining the best placement for large numbers of cameras within arbitrary building layouts. The method takes as input a 3D model of the building, and uses a genetic algorithm to find a placement that optimises coverage and (if desired) overlap between cameras. Results are reported for an implementation of the method, including its application to a wide variety of complex buildings, both real and synthetic.

Estimating camera overlap in large and growing networks

Large-scale intelligent video surveillance requires an accurate estimate of the relationships between the fields of view of the cameras in the network. The exclusion approach is the only method currently capable of performing online estimation of camera overlap for networks of more than 100 cameras, and implementations have demonstrated the capability to support networks of 1000 cameras. However, these implementations include a centralised processing component, with the practical result that the resources (in particular, memory) of the central processor limit the size of the network that can be supported. In this paper, we describe a new, partitioned, implementation of exclusion, suitable for deployment to a cluster of commodity servers. Results for this implementation demonstrate support for significantly larger camera networks than was previously feasible. Furthermore, the nature of the partitioning scheme enables incremental extension of system capacity through the addition of more servers, without interrupting the existing system. Finally, formulae for requirements of system memory and bandwidth resources, verified by experimental results, are derived to assist engineers seeking to implement the technique.

Tracking hand-off in large surveillance networks

This paper investigates the use of pairwise camera overlap estimates for supporting target tracking across large networks of surveillance cameras. We compare the use of camera overlap topology information to a method based on matching target appearance histograms, and also evaluate the effect of combining both methods. Tracking accuracy results are reported in terms of precision and recall for a 24 camera network. Camera overlap information is shown to deliver significant advantages for tracking when compared with simply matching target appearance histograms, due to its robustness to low quality imagery. We show empirically that this is the case even for automatically derived overlap estimates containing errors.

Empirical evaluation of the exclusion approach to estimating camera overlap

Making intelligent decisions on the basis of the video captured by a large network of surveillance cameras requires the ability to identify overlap between their fields of view. Without this information it is impossible to perform even simple analysis, such as distinguishing between repeated behaviours and multiple views of the same behaviour. Large-scale intelligent video surveillance thus requires a means of understanding the relationships between the fields of view of the cameras involved. The exclusion approach is the only method currently capable of performing online estimation of camera overlap for networks of more than 50 cameras, with a version of the algorithm applicable to 1000 camera networks having been published. Empirical evaluation of every such algorithm is critical to assessing its performance, and essential if comparisons between methods are to be made. This paper presents a method by which such an empirical evaluation may be carried out, and makes publicly available the data (including ground truth) on which it based in order that competing methods might be compared equally. Precision vs recall curves are reported for a series of experiments comparing the results of exclusion to ground truth. These results demonstrate the strengths and limitations of the exclusion-based estimation process, but show that the performance of the method exceeds the requirements of surveillance applications.

Ambassadors: structured object mobility in worldwide distributed systems

In many distributed systems applications deployed on the worldwide scale, it is latency, rather than bandwidth, that is the primacy determinant of performance. This paper describes Ambassadors, a communication technique using mobile Java objects within an RPC/RMI-like communication structure. Ambassadors minimise the aggregate latency of sequences of inter-dependent remote operations by migration of code to the vicinity of the server to execute those operations. Furthermore, because Ambassadors migrate within an RPC/RMI-like structure communication has well defined failure semantics, an important characteristic in supporting effective software engineering of distributed systems.

Unifying static and dynamic approaches to evolution through the compliant systems architecture

Support for evolution can be classified as static or dynamic. Static evolvability is principally concerned with structuring systems as separated abstractions. Dynamic evolvability is concerned with the means by which change is effected. Dynamic evolution provides the requisite flexibility for application evolution, however, the dynamic approach is not scalable in the absence of static measures to achieve separation of abstractions. This separation comes at a price in that issues of concern become trapped within static abstraction boundaries, thereby inhibiting dynamic evolution. The need for a unified approach has long been recognised but existing systems that attempt to address this need do so in an ad-hoc manner. The principal reason for this is that these approaches fail to resolve the incongruence in the underlying models. Our contention is that this disparity is incidental rather than fundamental to the problem. To this end, we propose an alternative model based on the compliant systems architecture (CSA), a structuring methodology for constructing software systems. The overriding benefit of this work is increased flexibility. Specifically our contribution is an instantiation of the CSA that supports unified static and dynamic evolution techniques. Our model is explored through a worked example in which we evolve an applications concurrency model.

Execution of VHDL Models Using Parallel Discrete Event Simulation Algorithms

In this paper, we discuss the use of parallel discrete event simulation (PDES) algorithms for execution of hardware models written in VHDL. We survey central event queue, conservative distributed and optimistic distributed PDES algorithms, and discuss aspects of the semantics of VHDL and VHDL-92 that affect the use of these algorithms in a VHDL simulator. Next, we describe an experiment performed as part of the Vsim Project at the University of Adelaide, in which a simulation kernel using the central event queue algorithm was developed. We present measurements taken from this kernel simulating some benchmark models. It appears that this technique, which is relatively simple to implement, is suitable for use on small scale multiprocessors (such as current desktop multiprocessor workstations), simulating behavioral and register transfer level models. However, the degree of useful parallelism achievable on gate level models with this technique appears to be limited.