Independent errors

Independent errors arise from random effects causing errors that manifest independence between any pair of measured values $m'$ and $m$, such that the error in $m'$ is in no degree predictable from knowledge of the error in $m$ (were that knowledge available). In an image, independent errors therefore arise from random effects operating at the image pixel level, the classic example being detector noise. The independent component of uncertainty (or, loosely, independent uncertainty) is the uncertainty contributed by the independent errors.