Improving Your CCTV Image

August 31, 2005 SSI Staff

Recently, the civilized world experienced new terrorist attacks with the bombings of the subways and buses of London. This was a stark reminder that the war on terror is far from over. In the aftermath of the bombings, the superior performance of the citywide London CCTV system proved invaluable in identifying those suspected of this horrific act.

As we all saw on the news, numerous arrests were made within days of the bombings due to identities verified on recorded video. This was a perfect example of why many refer to a CCTV camera and recorder system as the “silent witness.”

Swift action of the London bombing arrests was only made possible due to good quality video images. Many a time we see where a criminal act — such as a hold-up — is recorded and the video is either nonexistent or is of poor quality.

In recent years, poor video quality has become a greater problem due to the popularity of DVRs replacing VCRs. The digital video canvas has limitations security professionals need to understand.

Digital CCTV Book Recommended
In order to better understand the pros and cons of digital video, one must have good and current reference documentation. I recently reviewed the latest edition of an excellent book — “CCTV – Networking and Digital Technology” by Vlado Damjanovski, a true CCTV guru in the security industry.

Some of you old-timers may be familiar with Vlado’s original CCTV book, which came out in 1999 and has always been one of the best resources for CCTV technical information and applications. Vlado’s diversified descriptions and diagrams allow the reader to take their understanding of CCTV from basic how-to all the way to the mathematical explanations of a Fast Fourier Transformation.

Vlado introduced the second edition of his tome, which is published by Elsevier (Butterworth Heinemann, ISBN: 0-7506-7800-3), at the ISC West Las Vegas in April. The book is better than ever, and has been updated with new information on digital video, video networks and video testing. I strongly recommend it to anyone serious about understanding digital video security technology.

Video, Image Compression Methods
When working with digital video, you will see many references made to compression technologies. Simply stated, without compression techniques digital video would not be possible. The difficult challenge when working with digital video is deciding which compression works best with different applications.

If a camera was recording someone dealing cards at a casino, then a high image rate (live rate is 30 images/ second) would probably be needed. On the other hand, a DVR recording someone at an exit door could use a slower rate of maybe two frames/second for video motion detection.

There are basically two distinctive types of compression: video compression and image compression. Image compression is a 2D process affecting the horizontal and vertical dimensions. An example would be JPEG and wavelet (JPEG-2000) formats. Video compression is referred to as a 3D process that includes horizontal, vertical and time. This type of compression is called “temporal compression.” Examples are MPEG 1, 2, 4 and H.263/4 formats.

Video for Identification vs. Detection
One of the biggest challenges of selling CCTV is explaining to the customer what they can get for what dollar amount.

I strongly recommend you have a library of sample camera recordings on your laptop computer to help show views with different camera lens sizes and recordings at different compression rates to see what results will make the customer comfortable.

How much detail does the customer “expect” to see in the TV frame? Think of a CCTV digital TV frame as 720 X 480 pixels or you might hear the terms “full-frame” or “4CIF” resolution. By itself, CIF means one-fourth of the full frame resolution. Remember, a pixel is the smallest element on a digital video image.

Here are some guidelines:

To identify a person’s face (head = 75 pixels with body height = 480 pixels or 100-percent screen height)
To recognize a person’s face (body = 240 pixels or 50-percent screen height)
To serve as intrusion detection (body = 48 pixels or 10-percent screen height)
To serve as crowd control recognition (body = 24 pixels or 5-percent screen height)
To identify a license plate (plate height 24 pixels or 5-percent screen height)

Only Pay for What Can Be Seen
The preceding screen fill levels must be discussed and can be achieved by selecting the correct camera position and lens size along with recording resolution.

I can remember in the past when customers would invest in a high-resolution camera (460 TV lines) only to find out the recording resolution of a VCR was only 240 horizontal TV lines. While DVRs start off with a resolution of about 450 TVL, it does not pay to have higher resolution cameras being digitally recorded. Remember that these references are to digital video before compression.

In his book, Vlado nicely comments, “More attention should be directed to choosing a camera with a better signal/noise ratio, less smear or better dynamic range than to slight differences in high horizontal resolution (cameras) that nobody can see.”

Chart Handy to Test CCTV Systems
The questions I often get are: “Which DVR has the best image? How can I check them out?”

Ever since his first book, Vlado has offered those looking for such answers a video testing chart. This chart can reveal a large number of characteristics about CCTV video transmission, DVRs, cameras and monitors (see diagram on page 24 of September issue).

In the past, you had to set up a test area, with a camera, lighting and tripod to get a test image that could be viewed and recorded. A usable sample of the CCTV test chart is also on the back of the CCTV book.

In addition, Vlado now offers a small handheld electronic test pattern generator called TPG-8 (see photo above). This device is specially designed to electronically create a consistent test pattern chart. Understanding these charts can help analyze CCTV problems such as ground loops; excessive cable length; external EMF; bad quality from compression; and bad monitor, just to name a few.

The tester can be used to send test signals into various DVR inputs to see how well they perform under various compression configurations.

Additional information on the TPG-8 and the CCTV test charts is at www.cctvlabs.com.

If you enjoyed this article and want to receive more valuable industry content like this, click here to sign up for our FREE digital newsletters!