Deconvolution algorithms: optimize the Huygens deconvolution results.
Restore your microscopy images
The Huygens Software offers several restoration methods, optimized for specific imaging conditions.
Optimize the acquisition
A well-sampled image with no clipping is an important starting point for obtaining good deconvolution results. Click here for more imaging advise.
Huygens deconvolution in literature
Many scientific papers refer to Huygens deconvolution and its algorithms.
CMLE Classic Maximum Likelihood Estimation
The CMLE is the most general deconvolution algorithm available in Huygens, that provides good results for a variety of imaging conditions. You can optimize your CMLE results by fine-tuning the Acuity setting or, if needed, you can adjust the SNR value or number of iterations. The CMLE is fully GPU accelerated for even faster results.
Very robust, default algorithm.
Good results for different qualities of raw data.
GPU accelerated.
Confocal image courtesy of Leica microsystems CMS GmbH.
GMLE Good's Roughness Maximum Likelihood Estimation
The GMLE is a deconvolution algorithm optimized to handle high-noise images, typically obtained with STED or confocal imaging. Similar to the CMLE algorithm, you can fine-tune the GMLE results by adjusting the Acuity setting or, if still needed, the SNR value or number of iterations. The GMLE is also fully GPU accelerated.
Fast, less iterations than with CMLE.
Good results for high-noise images.
GPU accelerated.
GMLE Uses more RAM than CMLE.
STED image courtesy of Leica microsystems CMS GmbH.
QMLE Quick Maximum Likelihood Estimation
The QMLE is a deconvolution algorithm that gives good and fast deconvolution results on low-noise widefield images. In the QMLE you can tune the results by changing the Acuity value or, if needed, you can adjust the SNR setting or number of iterations.Fast, less iterations than with CMLE.
Good results for low-noise images.
Not optimal for noisy images.
Widefield image courtesy of Dr. Monica Pons, IBMB Madrid, Spain.
Suggestions for testing algorithms
We suggest to start with the default algorithm (+++) and then test also other algorithms (++ and +).
| CMLE | QMLE | GMLE | |
|---|---|---|---|
| Widefield | +++ | ++ | + |
| Confocal | +++ | ++ | |
| Spinning disk | +++ | ++ | |
| STED | +++ | ++ | |
| SPIM | +++ |
Iterative Maximum Likelihood Estimation (MLE)
Iterative Constrained Tikhonov Miller (ICTM)
Quick Tikhonov Miller(QTM)
Classic Tikhonov Miller (CTM), which is an improved version of Quick TM.
Widefield image courtesy of Dr. Livio Kleij and Martijn Vromans, Medical Oncology, UMC Utrecht, The Netherlands
Download: get started!
Test how the Huygens deconvolution algorithms can restore your images! Feel free to try out Huygens deconvolution and its other products by requesting free test licenses.
Download Now!