SAR Image Despeckling And Feature Extraction Based On Phase Information

SAR is a kind of high resolution imaging radar, the difference and relation between SAR and RAR are contacted each other, the difference and relation between the calculation of the resolution in the range direction and azimuth direction(Doppler Beam Sharpening) is described in detail. The special characteristics of SAR image lie on the geometry of radar imagery, the ratio of the electromagnetic wavelength to the real dimension of image pixel, the coherent & incoherent scattering, the polarization, the ground probing, the backscatter intensity of radar waves, etc and hence distinguishing itself from the optical image and others. Special matter and difficulties faced in SAR image interpretation including despeckling and feature extraction presenting as a world-wide problem to be overcomed are in the ascendant nowadays.In our research, the traditional despeckling and featureof edge and line extraction method and the characteristics of SAR image are studied, the local phase information of image in frequency transform domain calculated via double-tree complex wavelets are utilized, three algorithms are developed as follows: improved phase preserving algorithm for SAR image despeckling based on double-tree complex wavelets, improved dimensionless measurement of edge and line feature extraction algorithm from phase congruency calculated via wavelets for SAR image and dimensionless measurement of corner feature extraction algorithm from phase congruency moment analysis for SAR image. Corner feature extraction from SAR image is proposed for the first time in SAR image processing and analysis area. The majority of our work are as follows:· Mathematical bases of the three sorts of algorithms discussed above is described as follows: what is double-tree complex wavelets, how to analyze an image at different scales via wavelet, what is log-Gabor filters' important characteristics, how to obtain phase information via log-Gabor filters, how to...