Image denoising

Deep learning for fluorescence image restoration (denoising, deconvolution). Requires training on your data set but the procedure is described.

MorphoGraphX is a free Linux application for the visualization and analysis of 3D biological datasets. Developed by researchers, it is primarily used for the analysis and quantification of 3D live-imaged confocal data sets.

The main research interests adressed by MorphoGraphX are:

• Shape extraction
• Growth analysis
• Signal quantification
• Protein localization

SRRF is a high-performance analytical approach for Live-cell Super-Resolution Microscopy, provided as a fast GPU-enabled ImageJ plugin. SRRF is capable of extracting high-fidelity super-resolution information from TIRF, widefield and confocals using conventional fluorophores such as GFP. SRRF is capable of live-cell imaging over timescales ranging from minutes to hours.

A more modern approach for denoising / smoothening before segmentation, works like Gaussian blurring but preserves edges and boundaries. Listed in Fiji update sites. ## Algorithm Algorithm description is in [this page](http://www.ipol.im/pub/art/2011/bcm_nlm/) 2612. ## Example usage Localization of Membrane bound protein in Arabidopsis meristem was analyzed using the non-local-mean filter for refining its position 2613. ## impression It's effect is somewhere between Gaussian blurring and anistropic diffusion.

Variational algorithms to remove stationary noise. Application to microscopy imaging. This plugin allows to denoise images degraded with stationary noise. Stationary noise can be seen as a generalization of the standard white noise. Typical applications of this plugin are:

- Standard white noise denoising using a total variation and fidelity term minimization. Even though total variation denoising is not the state of the art (regarding SNR improvement), it may be very valuable for further tasks such as image seg- mentation).

- Destriping (the problem that motivated us to develop these ideas). 

- Deconvolution (even though most users won't be able to use this feature).

- Cartoon + texture decomposition which might be useful to compress images, analyse textures or simplify segmentation like tasks.