Maximum Likelihood Estimation
The iterative Classic Maximum Likelihood Estimation (CMLE) algorithm is a Restoration Method avaliable in the Huygens Software based on the idea of optimizing the likelihood of an estimate of the object given the measured image and the Point Spread Function (PSF). The object estimate is in the form of a regular 3D image. The likelihood in this procedure is computed by a Quality Criterion under the assumption that the Photon Noise is governed by Poisson statistics. For this reason it is optimally suited for low-signal images. In addition, it is well suited for restoring images of point- line- or plane like objects.
It is a Non Linear Iterative Method that allows the recovery of some lost information.
The CMLE method uses a Quality Criterion directly derived from concept of Maximum Likelihood: the I-divergence. It efficiently optimizes the quality criterion, and has the possibility to escape local minima which would lead to a wrong solution.
There are however situations in which other Restoration Methods come to front, for example when deconvolving 3D-Time Series, which is very computationally intensive. In this case you may consider to use Quick Maximum Likelihood Estimation-time (QMLE) which is much faster than the CMLE-time and will give excellent results as well. Furthermore, with the Good's roughness Maximum Likelihood Estimation algorithm (GLME) you will see improved handling of very noisy images, in especially STED and confocal data.
The time-enabled restoration tools are able to restore 4D Multi Channel images, provided the time option is included in the License String. If the time option is not present the tools will still handle multi channel images.
Confocal image courtesy of Leica microsystems CMS GmbH, deconvolved with Huygens CMLE algorithm.