Coupling optimal method of segmentation with restoration for target detection in SAR images

High-quality segmentation is there in a need in SAR images for appropriate target detection. In this paper, the features of Wiener filter are utilized with region of interest and Boundary based segmentation technique in SAR images. The discrete Wiener filter exhibits both linear and nonlinear characteristics. This filter compensates for the blurring due to the linear term and conserves the edges which are mainly used to distinguish the various objects with the gross representation of object of interest. The filter is defined both for edge preserving along with speckle noise cancellation in SAR images. A combination of features from two different types of multi-resolution and multi-channel filters, in general provides superior classification of SAR images. The experimental results show that the proposed boundary based and ROI technique segments the SAR image while the opted filter preserves significant features and noise removal. The fusion performance is evaluated on the basis of Mean square error, time, and signal-to-noise ratio.