基于双树复数小波的SAR图像噪声抑制
|
摘要
小波变换由于其良好的时频分离特性以及接近人类视觉系统的多分辨分析,在SAR图像的去噪和复原中得到了很好的应用,但是经典小波变换不具备平移不变性,且得到的高频分量的方向非常有限。复数小波变换是一种具有近似平移不变性、更多方向选择性且能够完全重构的双数正交小波变换,在图像去噪方面表现出更强的性能。建立了复数小波变换分解与重构的过程,并对分解后的实部和虚部图像的高频部分分别进行局部非线性软阈值法滤波。实验结果显示,复数小波变换较小波变换不仅滤除了更多的噪声,而且得到的图像边缘更加平滑。
The advantages of wavelet transform are that it can analyze signal in time domain and frequency domain respectively and the multi-resolution analysis is similar to Human Vision System(HVS).And now it is well used in de-noising and restoration of SAR images.But there are two disadvantages of DWT,lack of shift invariance and poor directional selectivity for diagonal features.Dual-tree Complex Wavelet Transform(DT-CWT) is a dual-tree biorthogonal DWT with approximate shift invariance,good directional selectivity and perfect reconstruction.DT-CWT is superior to DWT used in image de-noising.In this paper,the decomposition and reconstruction of DT-CWT have been constructed,and the real and imaginary parts of decomposed image have been filtered with local non-linear soft-threshold.Experiment results demonstrate that DT-CWT can filter more speckles and keep edges feature smooth.
The advantages of wavelet transform are that it can analyze signal in time domain and frequency domain respectively and the multi-resolution analysis is similar to Human Vision System(HVS).And now it is well used in de-noising and restoration of SAR images.But there are two disadvantages of DWT,lack of shift invariance and poor directional selectivity for diagonal features.Dual-tree Complex Wavelet Transform(DT-CWT) is a dual-tree biorthogonal DWT with approximate shift invariance,good directional selectivity and perfect reconstruction.DT-CWT is superior to DWT used in image de-noising.In this paper,the decomposition and reconstruction of DT-CWT have been constructed,and the real and imaginary parts of decomposed image have been filtered with local non-linear soft-threshold.Experiment results demonstrate that DT-CWT can filter more speckles and keep edges feature smooth.
引文
[1]肖国超,朱彩英.雷达摄影测量[M].北京:地震出版社,2001:54-56.
[2]Kingsbury N.The dual-tree complex wavelet transform:A newtechnique for shift-invariance and directional filters[A].IEEE DigitalSignal Processing Workshop[C].Bryce Canyon,1998.
[3]Nick Kingsbury.The dual-tree complex wavelet transform:Anew efficient tool for image restoration and enhancement[A].Proc.European Signal Processing Conference,EUSIPCO 98[C].Rhodes,1998:319-322.
[4]David Donoho L.De-noising by soft-thresholding[J].IEEETransactions on Information Theory,1995,41:613-627.
[5]赵池航,高成发,周百令.基于多重小波的海洋重力异常畸变消除方法[J].测绘科学技术学报,2006,23(4):265-268.
[6]张春华,许素芹,陶荣华,等.基于复小波变换的SAR图像斑点滤波方法[J].测绘学院学报,2005,22(4):269-271.
[7]Jalobeanu A,Blanc-F啨raud L,Zerubia J.Satellite image decon-volution using complex wavelet packets[A].Th埁me 3-Interaction hom-me-machine,images,donn啨es,connaissances Projet Ariana Rapportde recherche[C],2000:3955-4071.
[8]李强,王正志.基于小波分析的噪声抑制和数据压缩综合技术[J].计算技术与自动化,1998,17(1):35-38.
[9]胡召玲,郭达志,盛业华.利用小波变换抑制星载SAR图象的斑点噪声[J].中国矿业大学学报(自然科学版),2001,30(3):229-232.
[2]Kingsbury N.The dual-tree complex wavelet transform:A newtechnique for shift-invariance and directional filters[A].IEEE DigitalSignal Processing Workshop[C].Bryce Canyon,1998.
[3]Nick Kingsbury.The dual-tree complex wavelet transform:Anew efficient tool for image restoration and enhancement[A].Proc.European Signal Processing Conference,EUSIPCO 98[C].Rhodes,1998:319-322.
[4]David Donoho L.De-noising by soft-thresholding[J].IEEETransactions on Information Theory,1995,41:613-627.
[5]赵池航,高成发,周百令.基于多重小波的海洋重力异常畸变消除方法[J].测绘科学技术学报,2006,23(4):265-268.
[6]张春华,许素芹,陶荣华,等.基于复小波变换的SAR图像斑点滤波方法[J].测绘学院学报,2005,22(4):269-271.
[7]Jalobeanu A,Blanc-F啨raud L,Zerubia J.Satellite image decon-volution using complex wavelet packets[A].Th埁me 3-Interaction hom-me-machine,images,donn啨es,connaissances Projet Ariana Rapportde recherche[C],2000:3955-4071.
[8]李强,王正志.基于小波分析的噪声抑制和数据压缩综合技术[J].计算技术与自动化,1998,17(1):35-38.
[9]胡召玲,郭达志,盛业华.利用小波变换抑制星载SAR图象的斑点噪声[J].中国矿业大学学报(自然科学版),2001,30(3):229-232.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心