CameraCAD

CCTV calculator

DORI distance calculator

Enter a camera's resolution and lens; get the distances out to which it can detect, observe, recognise and identify a person, by the pixel densities of EN 62676-4.

Inputs

Units

Result

Need a set distance?

Search the camera database for models that meet a tier at a distance, best fit first.

at m
Find cameras

How the distances are worked out

DORI ties each task to a minimum density of pixels on the target, measured across the horizontal axis at the target's distance: detect 25 px/m, observe 62.5 px/m, recognise 125 px/m and identify 250 px/m. A camera with W horizontal pixels and a horizontal angle of view α spreads those pixels over a scene 2 · d · tan(α / 2) wide at distance d, so the distance at which it still delivers a density D is

d = W / (2 · tan(α / 2) · D)

Past about 120° a lens is a fisheye and no longer follows the pinhole model; there the calculator uses the angular (equidistant) projection, d = W / (α · D) with α in radians, the same rule the editor uses to draw a fisheye's rings.

Reading the result

The distances are the ceiling the optics allow in good light. Motion blur, compression, backlight and a face turned away all take detail off; a vendor's own DORI figures are usually computed the same way and are just as optimistic. For a varifocal lens, run the numbers at the setting you will actually install, not at the long end the datasheet quotes. DORI, explained goes through the standard, and Resolution versus detail shows why more megapixels on a wider lens can buy nothing.

Every model in the camera database has these distances worked out at both ends of its lens, and the editor draws them as rings on the floor plan, cut by the walls.