MADYTOOLS

DORI in CCTV: pixel density made simple

By Ali Madi · Updated

“Can this camera recognise a face at the gate?” DORI answers that question with one number — pixels per metre on the target. Here is how it works and how to calculate it.

DORI in CCTV: pixel density made simple

The four DORI levels

DORI comes from the European standard IEC EN 62676-4:2015. It defines how many pixels must fall on one metre of the scene (at the target distance) for each task:

LevelPixels per metreWhat you can do
Detection25 px/mSee that a person or vehicle is present.
Observation63 px/mSee what people are doing, clothing colour, general details.
Recognition125 px/mTell whether a person is someone you already know.
Identification250 px/mIdentify an unknown person beyond reasonable doubt.

How to calculate pixels per metre

Two simple steps: find how wide the picture is at the target distance, then share the camera's horizontal pixels over that width.

① Scene width (m) = Distance (m) × Angle factor
② Pixel density (px/m) = Horizontal pixels ÷ Scene width

The angle factor depends only on the camera's horizontal field of view — take it from this table:

Horizontal view110°100°90°80°60°45°30°20°
Angle factor2.862.382.001.681.150.830.540.35

For engineers: the factor is 2 × tan(angle ÷ 2), so Scene width = 2 × distance × tan(HFOV ÷ 2). The calculator above does this for any angle.

Example

A 2 MP camera (1920 px wide) with a 90° horizontal field of view, target at 10 m:

Practical rules

CCTV Designer Pro draws the DORI zones directly on your plan and shows the camera view at the real pixel density, so you can see the result before installation.

Plan a complete IP surveillance system on your own floor plan: camera positions, field of view, DORI coverage, a realistic 3D camera view, NVR storage and a ready-to-send report. Free, in English, Arabic and French.

Open CCTV Designer Pro →

Frequently asked questions

What does DORI stand for?

Detection, Observation, Recognition and Identification — four levels of image detail defined by IEC EN 62676-4.

How many pixels per metre are needed to identify a face?

250 px/m at the person's position, according to IEC EN 62676-4.

Is DORI distance printed on datasheets reliable?

It is calculated for ideal conditions; real installations need a margin for lighting, angle and compression.

Related guides