DECONVOLUTION REGULARIZED USING FUZZY C-MEANS ALGORITHM FOR BIOMEDICAL IMAGE DEBLURRING AND SEGMENTATION

Benoît Lelandais (1) , Frederic Duconge (1)

1 LMN - Laboratoire des Maladies Neurodégénératives - UMR 9199

We address deconvolution and segmentation of blurry images. We propose to use Fuzzy C-Means (FCM) for regulariz-ing Maximum Likelihood Expectation Maximization decon-volution approach. Regularization is performed by focusing the intensity of voxels around cluster centroids during decon-volution process. It is used to deconvolve extremely blurry images. It allows us retrieving sharp edges without impact-ing small structures. Thanks to FCM, by specifying the desired number of clusters, heterogeneities are taken into account and segmentation can be performed. Our method is evaluated on both simulated and Fluorescence Diffuse Optical Tomography biomedical blurry images. Results show our method is well designed for segmenting extremely blurry images , and outperforms the Total Variation regularization approach. Moreover, we demonstrate it is well suited for image quantification.

Keywords: Deconvolution; segmentation; deblurring; regularization; Fuzzy C-Means; quantification; molecular imaging

Domains: Image Processing [eess.IV]; Imaging

Benoît Lelandais, Frederic Duconge. DECONVOLUTION REGULARIZED USING FUZZY C-MEANS ALGORITHM FOR BIOMEDICAL IMAGE DEBLURRING AND SEGMENTATION. International Symposium on BIOMEDICAL IMAGING (ISBI): From Nano to Macro, Apr 2015, New-York, France. ⟨cea-01155379⟩