Research On Near Field High Resolution SAR Image Enhancement Method

Synthetic aperture radar(SAR),which has been widely used in various fields,is a high resolution microwave imaging radar.The near-field SAR high resolution imaging has became a hot reaseach topic which has great application prospect such as medical imaging,biological sciences,security inspection and anti-terrorist recently.However,some problems,especially high sidelobe and low contrast in SAR images,may bring serious quality issue and affect the subsequent application of near field high resolution SAR image.To be specific,high sidelobe may not only affect the target recognition but also form false targets due to the overlap,and low contrast may affect the image visual effect,making the image fuzzy.This thesis focuses on near-field SAR image sidelobe suppression technology and contrast enhancement technology about the problems of high resolution near field SAR image.The main research content is as follows:Firstly,this thesis analyses the image characteristics from the perspectives of near-field SAR image.The most important problems of the near-field SAR image are high sidelobe and low contrast.This thesis describes the resolution of image,the peak sidelobe ratio and the integral sidelobe ratio as evaluations for sidelobe suppression of high resolution SAR image and also proposed image entropy and enhance the degree as two evaluations for image contrast enhancement.These indicators provides an objective assessment for image sidelobe suppression and contrast enhancement research.This thesis makes the detailed studies for near-field SAR image sidelobe suppression technology.First,it expounds the mechanism of sidelobe,then it introduces the window function we usually used,at last,it points out the deficiencies about the window function for sidelobe suppression.This thesis research spectrum deformation method for sidelobe suppression by studying the relationship between the sidelobe trend and the image spectrum.This method is improved because of the phenomena about the disc broadening for image sidelobe suppression and the spectrum weighted sidelobe suppression method is proposed.It is proved very well by theoretical derivation.This method can restrain sidelobe effectively and strict control of the main lobe broadening,it is verified the effectivenes of the proposed method by experiment and comparison of multiple sets of data as well as the evaluation index of sidelobe.This thesis also makes the detailed studies for near-field SAR image about image contrast technology.First this thesis introduces some common approach for contrast enhancement and points out the insufficient about these methods.Then the image segmentation method based on 2D maximum entropy and maximum variance is introduced.Based on the image segmentation,the near field SAR images are enhanced,which overcome the serious problem of the background target interference in the near field SAR image,and makes the image contrast enhancement having a good result.The comparison and evaluation of the evaluation index using multiple sets of experimental data prove that the near field SAR image enhancement can achieve better results on the basis of image segmentation.