- Initially, pairwise comparagrams are constructed and unrolled, each assuming an arbitrary value of (e.g. ). The resulting function estimates for each pair of images will have the same overall shape, but each be scaled differently, apart from noise.
- Next these estimates of are used to estimate
the
*relative*values. Without loss of generality, may be assumed, and may be found. - Next the estimates of are consolidated (averaged together) by using the above estimates of to register them. This registered and averaged is now used to re-estimate the values, by comparing ratios: K_i = F^-1 - Q where is the logarithm of the reference quantity of light. (At this point, this first guess is often good enough to stop, but may undergo successive refinements by continuing as follows...)
- Finally, if desired, images are sorted in increasing order of exposure (least to greatest), and a final estimate of is made across all pairs of images, using the estimates of .
- If desired, this process may again be repeated, e.g. using that estimate to again estimate and so on.