Summary of the method

Easiest application

description

• If you stand up on a plane (ground) which is straight (but you can have a constant slope), take landmarks in this plane for the position of the near dof plane, focus plane and far dof plane. These will be represented with horizontal lines. You can place the focus plane in the mid in the image (not in the scene) of these 2 lines or closer to the near or far focus plane depending on what matters more. The mid position corresponds to a balanced position where foreground and background are equally important.

• These planes are landmarks to see the distant planes at different deep positions. You can imagine to extend this plane to represent other deep positions.

f# estimate

• If you choose a focus plane closer to the background:

  • This means that the the main contraints are for the foreground area to achieve minimum sharpness. The foreground area is delimited with the focus plane and near dof plane.

  • Estime the part of this area relatively to the image (1/3 of the image, 1/4) ... We call r this ratio.

  • Then the minimum f# that must be used is

f# > focal_length/(2*r)

For the foreground, just consider the background area instead.

adaptation of this rule

• the general formula is f# > f * (k/h) * r

k = 800 : This corresponds to an acceptable sharpness. You can adjust it but this value corresponds to the common circle of confusions used (but no need to change it with sensor size). In portrait mode use 1200.

h : height of the camera above the plane of reference. I use 1600 mm

This gives a ratio k/h = 1/2 but depending on your height, adjust it if this is really different.

If there is an important tilt down/up of the camera relatively to the plane, you have to take it into account, it will affect h.

You can have a look at the examples to see how this rule is applied.