Section: Software

nD-SAFIR: Image denoising software

Participants : Charles Kervrann, Patrick Bouthemy.

The nd-safir software (APP deposit number: IDDN.FR.001.190033.002.S.A.2007.000.21000 / new release 3.0 in 2012) written in c ++, java and matlab , removes additive Gaussian and non-Gaussian noise in still 2d or 3d images or in a 2d or 3d image sequences (with no motion computation) (see Figure 1 ) [4] . The method is unsupervised and is based on a pointwise selection of small image patches of fixed size (a data-driven adapted way) in spatial or space-time neighbourhood of each pixel (or voxel). The main idea is to associate with each pixel (or voxel) the weighted sum of intensities within an adaptive 2d or 3d (or 2d or 3d + time) neighbourhood and to use image patches to take into account complex spatial interactions. The neighbourhood size is selected at each spatial or space-time position according to a bias-variance criterion. The algorithm requires no tuning of control parameters (already calibrated with statistical arguments) and no library of image patches. The method has been applied to real noisy images (old photographs, jpeg -coded images, videos, ...) and is exploited in different biomedical application domains (time-lapse fluorescence microscopy, video-microscopy, mri imagery, x -ray imagery, ultrasound imagery, ...).

Figure 1. nd-safir software: denoising of a 3D image sequence in wide-field microscopy (GFP-Rab6A (Hela cell), UMR 144 CNRS Institut Curie)