Feature-enhancement of SAR images by Bayesian regularization

SAR imaging has an increasing interest in the surveillance and aircraft combat fields. The final aim is generally automatic target detection and recognition applications for assisted interpretation. To help these recognition processes it is important to get good quality SAR images, without loss of resolution. Particularly, we propose to enhance the image feature by reducing the sidelobe artefacts and smoothing the speckle. The methodology is the resolution of an ill-posed inverse problem by Bayesian regularization. The solution is then the value which minimizes a well-chosen penalised criteria. The originalities are minimization algorithm choice and real data applications.