The first civilian assignments in the field of Closed Circuit TV (CCTV) came at the end of the ’90s from municipalities, Schiphol, and the Dutch Railways. Much of the knowledge about surveillance cameras has been developed in a military context.
TNO’s experience with cameras is based, among other things, on its previous research into infrared equipment, photo cameras in aircraft and the like. The topic camera surveillance, however, concerns much more than technical knowledge: the design of observation rooms, the placement of cameras, the analysis of suspicious behaviour, the interpretation of group behaviour, the protection of data in relation to privacy protection, and the like.
TNO can fall back on the knowledge that it previously developed at IZF-TNO and its successor, TNO Applied Human Sciences (TNO Soesterberg).
After the first camera surveillance assignments in public spaces, a step-by-step plan was drawn up to be able to apply the available knowledge for this new task:
- The improvement of the camera image.
The images had to become sharper to better recognise people, objects, and vehicles.
- The analysis of the camera images.
This required programming of the cameras to highlight certain anomalies and odd behaviours. For example, when people walk to the right in the view of a camera, this is indicated. More complicated movements followed later. This required a lot of research.
The camera must alert the observer about certain detected behaviours and objects. The observer plays a crucial role in the total system. He/she must signal certain matters, such as deviant behaviour. The camera can support him/her in this task by sending a visual or auditory signal in the event of uncommon behaviour or behaviour that can be linked to crime or threat.
The observer must then interpret the identified behaviour. The latter is something that people can do well. However, the intention of TNO’s research is yo transfer that part of that skill into the algorithms in the camera. A new division of tasks between man and computer must arise. People can then concentrate on smaller details, while the computer provides a good overview and may recognise suspicious patterns. There is a list of around 196 behaviours that deserve the attention of the CCTV observer. This can range from nervous walking around to an elevated temperature of the person (certain advanced cameras can also measure the latter). TNO developed a programmed camera that recognises 48 activities through image analysis, such as throwing objects, running, hitting, kicking, and the like. If the camera detects such a movement, it gives an alert signal. The same may happen when certain sounds, such as shouting or screaming, are detected. It is even possible to combine these sounds for signalling with quick gestures or deviant behaviour.
To achieve the greatest possible safety with CCTV, a collaboration with various parties is necessary. This primarily concerns the security industry, i.e. producers of security systems, suppliers of observation rooms, developers of video content, and other stakeholders. Essential are the security guards who must learn to ‘collaborate’ with the cameras. Proper training of these employees is a prerequisite for the effective use of CCTV.
Camera surveillance often raises the question of its effectiveness. To what extent do cameras contribute to the safety of public spaces? That question can only be answered in conjunction with the entire security system. Camera surveillance also raises completely different questions. The relationship to privacy is a matter that regularly engages politics and the media, or the use of cameras in demonstrations and protests. This also applies to the possibility of abuse by totalitarian regimes. Such themes are also on the agenda at TNO.