Speckle Reduction And Real-Time Imaging Algorithm For Spaceborne SAR

Because the spaceborne synthetic aperture radar (SAR) could work day and nightunder all weather conditions and have the ability to penetrate the soil and thevegetarian, it has been widely used in many fields of national defense and economy,and become one of the most important earth-observation means. By the real-timeimaging processing of the spaceborne SAR, the ground can obtain the image of thescanning area quickly and make a fast response to the sudden events. In addition, thegood visual effect and the high-resolution are the goals of the SAR image. However,as a coherent imaging system, the SAR system generates the speckles inevitably. Thespeckles could degrade the resolution, blur the edge of the targets and hide the detailsof the image, and then influence the further image processing seriously.The speckle reduction method and the real-time imaging algorithm for thespaceborne SAR are studied in this paper. The main contents are as follows:(1) Aiming at the speckle suppression and the texture preservation of SARimages with low SNR, an adaptive speckle suppression method including twoprocedures is proposed. The wavelet threshold de-nosing method is first employed tofilter out most speckles and the image quality is enhanced primarily with higherimage contrast. In this operation, the homogenous area is selected adaptively tocalculate the threshold. Secondly, the FAB method, which is inapplicable to imageswith low SNR, is employed to smooth the residual noise and preserve the texturesimultaneously. the FAB method，different parameters are selected to calculate thediffusion coefficients for different scenes. The effectiveness of the proposed method isverified by the HJ-1C spaceborne SAR images.(2) In order to reduce the requirement of the high raw data rate on the generationrate of phase factors, the Chirp Scaling (CS) algorithm based on updating phasefactors regionally is studied. The effect of updating phase factors regionally on theSAR image quality is analyzed quantitatively. Finally, the effectiveness of the CS algorithm based on updating phase factors regionally is verified by the HJ-1Cspaceborne SAR real data processing.