---

Each face should be marked accordingly:

• category correct (А, В, C): Y/N;

• face correct (1,2,3): Y/N.

If category and face is correct score is: 3.

If only category is correct score is: 1.

If neither are correct score is: 0.

This should be completed for both faces.

A final score of 6 is the most desirable outcome - Pass.

A final score of 4 is an acceptable outcome - Pass.

A final score of 3 should be considered borderline and re-tested.

A final score of 2 should be considered borderline and re-tested.

A final score of below 2 is the least desirable outcome - Fail.

The VRN test has been designed to be a pass or fail test. If the observer is able to correctly identify all the characters displayed during the test then it is deemed that the system is suitable for human recognition of number plates. Otherwise the system cannot be considered to give an accurate result.

Table B.1 — Example auditor log sheet

Table B.2 — Example control room observer log sheet

Table B.3 — Example camera audit sheet

Location: Store A, Anytown Date: 15/6/09

Table B.4 — Blank auditor log sheet

Table B.5 — Blank control room observer log sheet

Table B.6 — Blank camera audit sheet

Location: Date:

Where VRN reproduction testing is required audit sheets similar to those above should be used.

В.12 Heads control sheet (for example only)

Figure B.1 — Heads control sheet

The test image can be downloaded on http://files.homeoffice.qov.uk/science-research/cctv test tarqets.zip

В.13 VRN control sheet (for example only)

The purpose of the text legibility test is to evaluate whether a CCTV system is capable of providing images that are suitable tor establishing textual details (specifically focused on vehicle registration marks). Test targets shall be used which provide letters with a similar shape like the following:

ouu ouv ovu

SPP SPR SRP

TEF TFE TFF

Figure B.2

The test consists of 9 segments of Vehicle Registration Number (VRN) characters. A random selection is presented towards the camera at an appropriate person, percentage, screen height, distance from camera.

An operator attempts to match the presented VRN to a reference list, thus determining the capability of the CCTV system.Annex C
(normative)

Test method of image quality
Guidance for the use of the video test target

The test method uses a standardised test target. It is used to evaluate the performance of a closed circuit television security system.

The test target (A3 format) is illustrated in Figure C.1.

The test target is easy to use for testing of coverage, determining acceptable image height, resolution, colour and contrast of the image. The test target is placed at strategic positions within the area of coverage as defined by the operational requirements or system specification, and detectability at each location is confirmed. This test should be carried out over the total light range over which the system is intended to operate.

As a basic principle horizontal and vertical angle between camera and test panel shall be smaller than 22,5 ⁰ in order to avoid optical distortion

Figure C.1

a)

The test image can be downloaded on http://vds.de/fileadmin/vds_publikation/ or http://files.homeoffice.gov.uk/science-research/cctv_test_targets.zip

If a differentiation of the big black and white segment peaks is possible, the quality level „inspect” is reached

.

I

b)

f a differentiation of the big black and white segment peaks is possible, the quality level „identify" is reached.

I

c)

f a differentiation of the big black and white segments is possible at the middle circle, the quality level „recognise" is reached.

d)

e)

f)

If a differentiation of the big black and white segments at the outer circle is possible, the quality level „detect" is reached.

6 colours can be differentiated: normal colour aptitude

NOTE Pink: Pantone 237 (Cyan 5 %, Magenta 50 %); Red: Pantone 485 (Magenta 95 %, Yellow 100 %); Yellow: (yellow 100 %);Green: pantone 360 (Cyan 60 %, Yellow 80 %); Blue turquoise: pantone 311 (Cyan 65 %, Yellow 15 %); Blue: pantone 285 (Cyan 90 %, Magenta 45 %)

6 colours can be differentiated: raised colour aptitude

NOTEApply a 50 % black filter on each colour from the first lin

e

g)

1

h)

1 grey scale values, deep black (background of test panel) and pure white

3 grey scale values, deep black (background of test panel) and pure white

Centimetre ruler for determination of the field of view

NOTE For German accident prevention regulation

j)

Notice for projection

1. Quality levels of images:

detection: 40 mm at real object relate to 1 pixel at least;

recognition: 8 mm at real object relate to 1 pixel at least;

identification: 4 mm at real object relate to 1 pixel at least;

inspect: 1 mm at real object relate to 1 pixel at least;

Pixel-oriented definition of the image quality: guarantee of a minimal quality of the image with no influence of the camera resolution, image-data-processing procedures etc. („one pixel remains one pixel")

Attention when compressing data, live-image and recorded image are to be compared!

2. in general especially the performance feature „identification" can be guaranteed only for a part of the camera surveillance range - see figure;

3. dead ranges depending of the mounting height are to be considered - see figure;

4. for cameras that can zoom and can be moved at least two reference images should be determined and documented.

Figure C.2

Dead range

Figure C.3

Annex D
(informative)

Guide to specifying CCTV parameters

One of the main failings of CCTV systems is that the owners and or installers do not have a clear idea of the purpose of each camera and the level of detail needed to achieve that purpose. Cameras that are attempting to perform too many functions or have no clear purpose are a waste of resource as they are unlikely to produce useable images. For these reasons it is highly desirable to have a clear Operational Requirement (OR) not only for the site in general but also for each individual camera. This need not be as daunting a task as it may sound if the procedure outlined below is followed.

It is suggested that a modular approach is taken to specifying CCTV, as common area types and risks exist across a broad range of environments. The table below contains examples of these building blocks with minimum image quality and frame rates dependant on the perceived risk level. The image quality (image size and compression level) consistent with each of the categories should be defined and tested using a suitable image quality test. The frame rates given are suggested minimum values. For those areas marked with an asterisk (*) it is acceptable that the default frame rate could be reduced if an alarm trigger mechanism is deployed that would cause the frame rate to increase on activation. Any system that relies on frame rate increasing after alarm trigger should buffer the video so that several seconds before the alarm trigger is captured at the higher frame rate.

Table D.1 —Suggested CCTV building blocks (1 of 2)

Table D.1 —Suggested CCTV building blocks (2 of 2)