printer friendly version

The human factor side of video management systems

A video management system (VMS) is central to, and the most vital element to any control room operation using CCTV as part of its service delivery. All too often, it is seen as a technical solution rather than as an operational solution. The scope of the VMS, the range of capabilities, handling of recording and retrieval, the sophistication of its handling of events, and reporting capabilities are all important things that come with the system. However, strategically, the human factors issues in how it is implemented, the ease of use, and the suitability to enhance operator functions are fundamental and critical issues in how successfully it works.

I have said in the past that the way cameras are viewed and displayed as part of a strategic approach to surveillance is the basis for effective detection of incidents and situational awareness. This display of cameras through the VMS interface will be influenced and constrained by the physical components of the control room, including control room design, placement and available space on monitors – you can only display according to the available viewing space available to you.

Ideally, the hardware and software components of the control room system should be aligned and complimentary to your surveillance strategy from the beginning. This includes dealing with potential camera time ‘drift’, where different cameras’ date/time stamps deviate from the accurate time. This should be addressed through network design and configuration at a server level, avoiding time stamps that differ from each other and the original incident time. The equipment should allow the VMS to configure and display its optimal capacity and contribute to control room service delivery.

Seeing what needs to be seen

Viewing layouts are often set during the initial installations of the VMS. Often, this is done with too little appreciation of the viewing strategies to which these layouts will be put. One of the key capabilities of a VMS is to allow flexible configuration and saving of camera layouts, as well as changing things on the fly if an incident situation warrants it. In some cases, changes to the VMS configuration can only be done by technicians. I have emphasised repeatedly that different cameras are more important at different times; displays should feature this and have the capacity to be easily changed.

Incident conditions such as tracking suspects, monitoring large group dynamics, or reviewing incidents in particular, may also call for different groupings of cameras. It should be possible for any modern VMS to allow an authorised operator to configure a group of cameras to be displayed in a way that the movement or activities of a person can be easily tracked to the camera next to the initial one displayed, either on a live basis or during a review. Even high-end VMSs sometimes limit the ability to position cameras intuitively relative to each other in a grouping, resulting in the kind of situation where somebody in a camera view is moving right, and the movement in the next camera is shown in the top left corner or somewhere else obscure.

Also, different operators may differ in what they are assigned to watch or have preferences about setting up their viewing layout. With some guidelines in place, they should be allowed to optimise camera layouts according to their specific duties, understanding or preferences, if these align with delivering results. Macros to change to designated layouts, saved layouts for specific purposes or particular personnel, and even automatic layout changes according to alarms, times of day or risk conditions are desirable within a VMS.

Another severe limitation when setting up VMS systems is camera naming conventions. I have seen camera naming that looks more like complicated password codes, which makes selecting cameras difficult. Camera naming needs to be done according to common sense, relevance to the different areas being viewed on site, in order to easily switch to these cameras on a logical and flowing basis. Camera names also should be able to be edited, and not just by technicians. Some conventions should still be applied to this to ensure there is no chaos with uncontrolled numbering.

Cameras usually maintain a numbered status, and camera numbering must also be considered according to area. Control rooms also need to consider that additional cameras are likely to be added to the system, as this often results in wild fluctuations in camera numbers associated with an area, as new cameras are included.

Camera names are also displayed on the video feed, which has implications when shown to other people, including at legal proceedings and as court-presented evidence. If you say a camera is covering a particular area, and yet the camera name does not bear any similarity to the location of the incident being presented in court, it is not unexpected for people in these situations to query whether that camera is viewing the affected area. This may be due to the naming of the camera for very relevant reasons, but it still presents a mismatch and comes back to a common sense naming convention for your cameras.

The camera list displays on VMS systems also often leave something to be desired, with long lists that need to be scrolled down to access cameras; with little organisation or grouping. Where you are working with 200, 500, or more cameras, this can get onerous and lengthy at best. Initial set-up should consider both naming conventions as well as how to incorporate these and groupings of cameras according to areas into sensible and easily used camera listing. Names should also be viewable – for example, in some instances, there is an identifier code followed by the camera name. This sometimes cuts off the full camera name from the screen list and sometimes means that the list window border has to be dragged to the side, obscuring camera viewing areas. Camera listing is an important part of the display and operator actions, and reconciling these is not always easy.

Effective integration of analytics and AI

Video analytics or AI capabilities are increasingly being delivered as part of VMS capabilities. The cost/benefit of the analytics contribution and its sophistication should be considered when deciding on a system. This includes ease of setting up on-camera analytics and support on this. Where the VMS or an integrated independent analytics product has built-in video analytics or AI capabilities, responses to alerts and alarm conditions should be seamless to the continued operation and not detract from core viewing duties.

Many VMSs have required operator response logging for an alert, to confirm it has been viewed and addressed. This could interfere with normal monitoring, particularly in the case where analytics are generating large numbers of false alarms on the system, and the logging process is unnecessarily complicated.

One of the most underrated but one of the most critical areas in terms of human interface is the review and search capability of the VMS when looking at past incidents. The ability to quickly access the footage in the required time period, to view it in various ways, from slowed motion to reverse viewing, and to save or capture video or frames is an essential part of reviewing and evidence collection. I have found many of these VMS review interfaces clumsy and not intuitive to use. Yet, they provide the basis for how the VMS recorded data will be used to follow up, review and present information for evidence purposes. Along with the other features, I would make this review interface a vital part of any decision to choose a VMS.

VMS features may provide huge potential for CCTV usage and data management. However, the human factor side of these systems is often neglected when it comes to actual implementation. It calls for strong surveillance management team input into the installation and setting up process in line with the surveillance strategy. It also needs constant feedback and regular performance reviews to check that supervisors and operators can use feature sets to quickly and effectively achieve results. One size does not fit all in this sector, and the ability to have a flexible system that can be adjusted to needs and easily customised by users is an important consideration in the choice of any VMS.

Dr Craig Donald.

Dr Craig Donald is a human factors specialist in security and CCTV. He is a director of Leaderware which provides instruments for the selection of CCTV operators, X-ray screeners and other security personnel in major operations around the world. He also runs CCTV Surveillance Skills and Body Language, and Advanced Surveillance Body Language courses for CCTV operators, supervisors and managers internationally, and consults on CCTV management. He can be contacted on +27 11 787 7811 or [email protected]

Share this article:

Further reading: