Understanding CCTV

How is a Camera’s TVL Determined?

Television Lines (TVL) is a measure of the quality of a camera and is usually used to

describe an analog system. TVL is measured in lines of horizontal resolution (the number

of alternating light and dark lines that can be resolved in the X-direction).

Traditionally, the measurement of television lines for an analog system is in lines per

picture height. This is the number of television lines that can be measured in the largest

square (1:1 area) of a screen.

For example, the active window of a PAL signal is 720/704 x 576 depending if it is DVD

(720) or TV (704) quality but the maximum TVL for this signal can only be 576 since

this corresponds to the 1:1 area. This is shown in the picture below. The black rectangle

represents the 720/704 x 576 picture while the red square shows the 1:1 area of the

picture. This area only has 576 pixels in the X-direction.

Note that even though a camera may output a PAL TV signal, its TVL can be

considerable less than 576. TVL is a product of many factors including the sensor size

and the quality of the camera’s lens.

A standard method of measuring TVL is by the use of a video resolution pattern such as

the EIA-1956 pattern shown below.

Using a proper set-up, the TVL of a camera is determined to be the point at which the 4

triangular lines converge into 3 lines.

What does TVL mean in a CCTV system?

TVL is a representation of the quality of an image. As such, this parameter can be used to

properly pair a camera with a DVR.

A DVR digitizes an analog signal and stores that information on a hard disk drive. There

are 3 parameters that can be used to quantify this process. They are:

• Analog-to-Digital conversion resolution (ADC resolution)

• Compression algorithm

• Compression bandwidth

The ADC resolution is the number of pixels (data points) that are used to represent the

analog signal. The more pixels that are used in this process, the better the representation

that the digital image will have to the analog signal. Common resolutions are:

• CIF 352 x 288 pixels (288 maximum TVL)

• 2CIF 704 x 288 pixels (288/576 maximum TVL*)

• 4CIF 704 x 576 pixels (576 maximum TVL)

• D1 720 x 576 pixels (576 maximum TVL)

*Some people might argue that this is really 576 TVL since this resolution can capture 576 TVL in the

horizontal direction (even though the vertical resolution is limited to 288). Recall that TVL is defined as the

number of alternating light and dark lines that can be resolved in the X-direction.

The ADC resolution will determine the maximum TVL that a CCTV system can represent.

The compression algorithm MJPEG 4 H.264etc.) and bandwidth (500 kbps, 1 mbps,

etc.) determine how much disk storage is used to store the digital image. The better the

compression algorithm or the higher the bandwidth, the closer the digital image will be to

the analog source.

The compression algorithm and bandwidth will determine the actual TVL that a CCTV

DVR can capture. This will almost always be less than the maximum TVL since the

compression algorithms used are lossy (information is lost in the compression of the

image).

The ADC resolution, compression algorithm, and bandwidth of a DVR should be chosen

to properly match a camera’s TVL number.

For example, it makes little sense to purchase cameras that are capable of 540TVL and

then use a DVR that can only record CIF images at a very low bandwidth. CIF has a

maximum TVL of 288 which can easily decrease to below 200 depending on the

compression algorithm and bandwidth used.

How to measure the TVL of a DVR?

To measure the TVL of a DVR, a test pattern should be recorded and then played back on

a high quality monitor. The TVL can then be measured by noting the point at which the 4

triangular lines converge into 3 lines.

Note that the recorded and not the live image must be used to determine the TVL of a

DVR.

Even though a high quality camera can be used as the test pattern source, this could easily

lead to discrepancies since test set-up might not be 100% reproducible due to differences

in lighting condition, camera position, and so forth.

Another approach would be to use a pattern generation such as the one available from

CCTV Labs (http://www.cctvlabs.com/testpatterngenerator.html). This pattern generator

comes pre-programmed with various patterns that can be used to measure the quality of a

DVR.