Blind multichannel image deconvolution with regularization
Wided Souidene Mseddi, Karim Abed‐Meraim, Azeddine Beghdadi · 2005
Blind image deconvolution is a very active research field. It has been proved to be useful in many situations related with image processing: microscopy imaging, remote sensing and astronomical imaging. Several methods were developed in a single-channel framework. Unfortunately, these methods encounter some problems even in noiseless case. Recently, a Mutually Referenced Equalizers (MRE) method was proposed in a multichannel framework which ensures perfect Point Spread Functions (PSF) recovery in the noise free case. In this paper we propose to extend the MRE method to the noisy case by means of a regularization procedure. We also evaluate the restoration performance using an objective image quality assessment measure.