As the revolution of IP and HD CCTV becomes part of everyday life for retail and corporate business, it is more important than ever that the fundamentals of basic operation and system objectives are met. The alternative is a very expensive electronics array that provides little useful information to prosecute an offender.
As an example, and to highlight the issues related to providing useful CCTV, the following scenario will be of benefit when designing, installing and maintaining a surveillance system.
A few years ago, at a licensed premises, which included a gaming facility, there was a confrontation between two individuals resulting in the death of one. Almost all of the activity was captured successfully on an analogue DVR (Digital Video Recorder). The DVR was recovered and removed from site by police and a technician. To fully secure the evidence, the unit was stored in a secure vault within the court complex.
At the time, we suggested to police that a backup be made of all the video data for the full 24 hour period surrounding the incident from all 16 cameras on site in case of a hard disk failure while the unit was in storage. Following consideration by police and the court, a copy of the video data was transferred onto a new hard disk drive supplied by police.
During the process of backing up video data, the supervising officer (also the officer that attended the venue on the night of the incident) asked how best to ensure that the DVR would be fine and ready for use in court when the need occurred. This posed a number of issues.
The question was raised regarding how long might it take before the evidence on the original DVR could be presented. The answer was understandably vague as the processes involved in preparing the case, which involved both police and the court, were very detailed and would take considerable time to complete. The main issues surrounding the preservation of the DVR were as follows:
- That the unit not undergo any significant or unnecessary movement
- Under no circumstance could the unit be dropped during relocation,
- Neither should the unit be opened in any way
The last point bit triggered a bigger issue that had been considered previously.
All the people involved in the handling and storage of the DVR to this point were supremely confident that the data stored on the original unit was intact, it was encrypted and watermarked in such a manner that would be acceptable for use in a court environment. They could also be reasonably confident that the unit would not need to be moved or relocated and certainly not dropped. The unit was tagged appropriately, boxed with its polystyrene packaging and stored securely….so “all should be well when we get to court,” we all said with a sigh of relief. We shall return to this particular incident in a moment.
After a couple of weeks had passed, a shoplifting incident occurred at an un-related site, and video data was backed up from the ageing DVR and provided to police. It was noted during the process of completing the on-site back up that although the incident had occurred during daylight savings time, the DVR had not automatically updated its time correctly as the time server IP address in the DVR was no longer valid. Upon further investigation, it was found that the CMOS battery on the mother board was dead resulting in the time and date on the unit being reset to the unit’s default, most likely following a complete power re-boot.
This initially triggered some concern about the ability to retain secure video data on older models of DVR. However, further thought lead us to explore the necessity of ensuring that systems were maintained in a manner that would reduce the risk of unreliable video evidence. This incident immediately prompted concern regarding the previously discussed DVR, which was being held for evidentiary purposes by the court, with my immediate thought being “what if”; a concern which would very soon become pertinent.
Contact was made with police, to point out that if the DVR in question was going to be held in a secure evidence store for an extended period of time, the CMOS battery might fail which would not corrupt any of the existing video data on the drives, but could and probably would reset the system time to the default 01-01-2000. Furthermore, on power up, the unit would begin recording from that date. Therefore, surely when the unit was used in court to provide evidence, the recorded video history would clearly show that the most recent video data recorded appears to be up to 8 years older than the video being presented for evidence. Did this render the video evidence of the incident unusable or unreliable?
It was decided that during the lead up to trial, the DVR would be periodically brought out of the evidence store and run up in a secure environment within the court facility with a police and court witness present at all times. This would ensure and confirm that the DVR would be functional and ready to present the video data recorded at the time of the event from its original source not a back up. It was also decided that, during the trial, I should be available to present factual evidence with regards to the construction of the DVR, the software used to create the recordings and original installation, maintenance, recovery and continued care of the unit.
The positioning of cameras on site was considered at the time of installation and of course, had evolved during the life of the CCTV system to reflect the growing needs of the venue. We had discussed during the installation that the balance between the length of recorded history and the number of frames each camera would capture each second was a critical decision, but could be easily adjusted over the initial month to get the best possible outcome. The eventual frame rate agreed upon was 7 fps (frames per second). I felt that this would provide around 60-70 days or recorded history, and all at a very good motion sensitivity.
The cameras utilised were of relatively good quality with some even being re-used when the system was upgraded. The cameras were at least 500TVL day/night, so good images were produced, regardless of the local lighting and weather conditions. A number of the cameras were inside the building and a couple were on the external façade to cover car parks and entry points.
The Trial:
The trial ran over a number of day’s with my attendance being required for the duration of the hearing. Although I had attended many court cases to provide support to police or the court, this was the first that involved t
he death of a person clearly recorded
on CCTV.
Coincidentally, the camera that captured the final event was not an expensive 600TVL day/night camera with sensup or clever BLC adjustments; it was an older, full bodied camera in a dome housing that had been installed some years before.
My initial court attendance involved being required to re-install the DVR into the actual courtroom where the trial would commence later that day, and then train prosecution and defence barristers on the use of the unit’s playback characteristics. I was present when the incident was shown to the court and was then questioned at length on the construction, programming and security of the recorded video data. I was also asked about my personal experience and longevity in the security industry including my historic knowledge of this particular DVR and software. All of the aforementioned questions where answered with little need for further questioning. However, it should be remembered that when giving such evidence, one must only state the facts and no assumptions may be made regarding any part of the evidence provided, and to remain calm and confident in your answers.
I would like to say at this point, that the idea of having to provide evidence in court is an easy thing to do, but it isn’t. It is stressful and sometimes disconcerting. The most difficult points to relay to the court were the understanding of video compression (why does it need to be compressed), frame rate (explaining what might have been missed in the other little bits of that one second period recorded), motion detection (what bits do not get recorded) and the watermarking of video (security), all of which would have a direct bearing on the final outcome of the trial. To make these points clearly and concisely in a manner that the court could understand, a large white board came in very handy.
The outcome of the trail was never going to be positive for the individuals involved but it proved that the best possible CCTV evidence is a critical component for police and the court to come to a definitive decision.
As a result of my involvement with this and other court cases, I have been able to develop a number of key points which need to be considered when implementating and maintening any CCTV system. They are as follows:
- Regularly maintain the system including camera mounting, cleanliness, focusing and alignment to the subject.
- Check the system time against a known correct source frequently and correct accordingly.
- Records of system maintenance must be retained.
- Only use recording equipment that records watermarked images and, therefore cannot be manipulated, altered or changed in any way.
- Restrict the number of employee’s that have access to the CCTV equipment.
- Provide training to key staff on the use of the system ensuring that they can back up video for police on request.
- Remember that the person that provides the video data to police will become a witness.
They should have a good understanding of the CCTV system, and be able to give evidence of the steps they took to make a backup copy of the video data supplied to police.
Some guidance of camera location, purpose and objective is provided below in accordance with SAPOL (South Australian Police) requirements for Closed Circuit Television.
