Time stand still

In security applications, it’s common with an analogue recording system to add a multiplexer which then allows up to 16 cameras to be recorded to one VCR by processing the images and sequentially laying them down on to the tape... Camera 1, Camera 2, Camera 3, etc up to 16, and then round again... Camera 1, Camera 2, Camera 3, etc. It’s also common to slow down the tape speed (time lapse) so that the tape which would ordinarily run for three hours at normal speed is slowed down by a factor of eight times such that it runs for 24 hours.

In all honesty, this is convenient because it means one tape covers each 24-hour period. However, it should be noted that the interval between images is also slowed down so, in 24-hour time lapse mode, that interval is approximately 3-5 seconds. To obtain an improved frame rate, it’s necessary for the end user to decrease the number of cameras and/or speed up the tape.

There are other considerations – including tape wear and pick-up head wear – that demand regular tape replacement and machine servicing. Additionally, full compliance with recommended practice for magnetic media includes degaussing (ie magnetic erasing) the videotape before each pass. This can turn out to be both a costly and time-consuming business.

Digital recording has been nothing less than a revelation... not to say a revolution. Many systems have appeared on the market these past few years, with a vast array of kit on display at IFSEC 2006. The benefits over and above analogue recording are considerable... no tape changing, no (or very low) maintenance and hardly any management time all help to reduce system costs.

Operationally, digital recorders are also friendlier in that different cameras may have different recording profiles (it is possible to retrieve images while the machine is still recording, and to search far more easily for previously recorded footage).

A major benefit with larger systems is their ability to connect to a network – enabling remote searching, viewing and reviewing.

Time lapse or event recording?

Traditionally, the recording schedule using analogue/multiplex combinations was continuous. There was no other way. You either recorded everything or nothing. Digital video recorders (DVRs), though, possess the ability to start and stop recording very quickly. Due to this, it’s possible to record only when there’s activity and, thus, use your hard disk storage capacity more economically.

In-built movement detection is most commonly used, but any kind of alarm input is possible. It should also be feasible to mix continuous recording and event recording using time scheduling (for instance, continuous during the day and then movement detection at night and weekends, etc). It should also be viable to have one recording speed for time lapse, then an option to speed up to faster in alarm mode, etc.

Take this simple example... Blanket recording for 24 hours per day for one whole week equals seven days times 24 hours which equates to a total of 168 hours. More selective recording? You could record continuously during office opening hours of, say, 9.00 am through to 5.00 pm from Monday to Friday, and then revert to event recording for the rest of the time. Five days at eight hours per day equals 40 hours plus some night recording. That would give you a total of around 50 hours, thereby saving 118 hours of recording space.

In terms of storage time for images, historically security professionals have stored for one month as standard (from 28 to 31 days). Notwithstanding many statements in sales literature, there is still no legislation which specifies this. The Data Protection Act 1998 does NOT specify a minimum of 31 days’ storage. End users store CCTV images for as long as they need – no more and no less. There are often good reasons for storage over a longer time period, but then again there are equally good reasons for not doing so. Indeed, many commercial organisations use a much reduced period of seven to ten days.

The Data Protection Act 1998 does NOT specify a minimum of 31 days’ storage. End users store CCTV images for as long as they need – no more and no less. That is the reality of the situation

Hard disk or digital tape?

There’s little doubt that hard disk has many benefits over tape. It requires no service, no maintenance and no tape removal, storage or management. Hard disk also provides very fast access [to images]. Reductions in unit prices coupled with increasing capacities have, in recent times, rendered these systems very affordable for the end user.

DVRs employ a method known as ‘video compression’ to ‘squeeze’ the images into small packages for more economic storage. This is a process whereby the picture information is replaced by a more compact mathematical description and placed onto a storage media. Decompression is the reverse process (pictures are re-built pixel by pixel).

Several different compression algorithms are available on the DVR market. They can be very difficult to understand. Some systems claim to record all cameras virtually in real time (up to 25 pictures per second), while others require slower recording rates. Some systems use very little disk space while others use a whole lot more for the same recording.

To try and put these conflicting pieces of information into perspective... The compression methods influence both the theoretical maximum speed of recording and the amount of storage needed. The commonly used compressions in digital recording systems are MPEG, H261, Wavelet and JPEG. They are each very different in operation.

The Moving Picture Experts Group (hence MPEG) is a Working Group of the International Standards Organisation (ISO) tasked with the development of standards for the coded representation of both digital audio and video. Established in 1988, MPEG technology operates by way of what is commonly called a ‘refresh’ type scenario. This method does not compress each individual picture.

A full image will be taken periodically. These individual frames are called Intra-Frames, or I-Frames. Those images occurring between I-Frames are not transmitted in full, only the changes. These are called P-Frames.

MPEG 1, 2 and 4 explained

MPEG 1 is the standard on which such products as Video CD and MP3 are based. MPEG 2, on the other hand, is the standard on which digital television set top boxes and DVD are based.

Different again is MPEG 4. This is the standard for multimedia for the fixed and mobile web. In essence, it’s a high speed data transmission compression method originally designed for audiovisual applications which is now being widely employed in Internet Protocol (IP)-based surveillance solutions. It operates largely on the basics of MPEG.

H261/H320 were originally designed for videoconferencing. This method takes a reference frame comprising the whole picture (reduced size) which is then ‘refreshed’ by updating changes (then taking another reference frame after a period of time). By contrast, Wavelet technology is a method which allows high compression to smaller file sizes, but uses approximating functions when reconstructing the image.

Despite what all of the manufacturers’ glossy brochures may say, the only real way for the end user to evaluate a system is to look at it and decide for themselves if the image quality and recording speed are suitable for their needs

Many of you will be familiar with JPEG. This is short for the Joint Photographic Experts Group, the name of the Committee that wrote the standard. JPEG is configured for compressing full-colour or greyscale images designed for still images. Each picture is captured as an individual ‘snapshot’ of the complete picture, in essence as a series of ‘stills’. There is no inter-frame compression. M-JPEG4 is a video adaptation of the JPEG standard. It simply treats a video stream as a series of still photos, compressing each individually with no inter-frame compression and finally presenting them as a continuous video sequence.

CCTV image quality

The image ‘quality’ is governed by the structure of the picture file expressed in the number of pixels (or picture ‘points’) making up the image. Internationally-recognised image file sizes are CIF, 2 CIF and 4 CIF. Image ‘quality’ is also determined by how much data is stored in each picture. This ‘amount of data’ is usually measured in Kilobytes (1 Kilobyte is equal to 1,000 bits of data).

File sizes may be increased or decreased as a result of adjusting the system – which impacts directly on the speed of recording, the size of the database required and, ultimately, the ‘quality’ of the image.

Despite what all of the manufacturers’ glossy brochures may say, the only real way for the end user to evaluate a system is to look at it and decide for themselves if the image quality and recording speed are suitable for their project needs.

On that note, while demonstrations are useful, remember that certain systems aren’t very tolerant to large amounts of movement in the picture. Others will record at high speed, up to and including real time, but is that really necessary? In many cases a slower recording speed but higher resolution is probably more suitable for the end user.

In other words, there is no ‘one size fits all’ system when it comes to specification. You must think carefully about what you want to achieve before shopping for your system. Although the Home Office Scientific Development Branch (HOSDB) has been looking at many of these issues, there have been no official comments made to date.

Backing-up your system

It must be possible to back-up to an easy-to-use removable media. Ideally, it should be feasible to back-up a given incident at a single position, not back-up each group of incident pictures from each individual machine if there are more than one involved on the system. This is fine for simpler installations, but any networked installations should be designed from the beginning as a server platform to allow the seamless networking of multiple machines.

Once the back-up is completed, it must be possible for it to be easily read off-site by, for example, members of the police service. The method suggested by the HOSDB is to back-up to CD-Rom or DVD with any viewing software appended during the process so the images rendered can be easily viewed or reviewed on a PC without the need for any special software.

While DVD offers higher capacities, there are potential compatibility risks with the viewing PC. With CD-Rom there is a single standard, but with DVD there are several.