swMATH ID: 30680
Software Authors: Serge Dmitrieff, François Nédélec
Description: ConfocalGN: A minimalistic confocal image generator. Validating image analysis pipelines and training machine-learning segmentation algorithms require images with known features. Synthetic images can be used for this purpose, with the advantage that large reference sets can be produced easily. It is however essential to obtain images that are as realistic as possible in terms of noise and resolution, which is challenging in the field of microscopy. We describe ConfocalGN, a user-friendly software that can generate synthetic microscopy stacks from a ground truth (i.e. the observed object) specified as a 3D bitmap or a list of fluorophore coordinates. This software can analyze a real microscope image stack to set the noise parameters and directly generate new images of the object with noise characteristics similar to that of the sample image. With a minimal input from the user and a modular architecture, ConfocalGN is easily integrated with existing image analysis solutions.
Homepage: https://www.sciencedirect.com/science/article/pii/S2352711017300444
Dependencies: Matlab
Keywords: Synthetic image; Image analysis; Bioinformatics; MATLAB; SoftwareX publication; confocal image generator
Related Software: Huygens; SMeagol; BlurLab; DeconvolutionLab2; ilastik; Matlab
Referenced in: 0 Publications

Standard Articles

1 Publication describing the Software Year