Wavelet Transform-Based Approach for Interferometric SAR Image Noise Reduction
-
摘要: 根据干涉合成孔径雷达(SAR)图像的噪声来源和性质以及干涉合成孔径雷达信号处理的特点,选择了具有对称性和紧支性的双正交小波变换应用于干涉SAR图像的噪声抑制,提出了基于小波标架表示的干涉SAR相位图像的降噪算法.与传统的低通滤波器及中值滤波器处理的结果对比显示,利用小波变换方法处理干涉合成孔径雷达图像以提高干涉相位精度的方法是可行的,并且小波变换方法是基于观测数据的自适应算法.Abstract: Thermal and "Speckle" noise is an obstacle to obtain high accuracy phase in interferometric SAR images. According to the main sources of noise in interferogram, the noise character and the characteristics of interferometric SAR processing, biorthogonal wavelet transform with symmetry and compact support is introduced into the reduction of noise in interferometric SAR images. Based on the wavelet frame representation, a new algorithm for noise suppression in interferogram is proposed. Compared with the traditional low pass filter and median filter, it is showed that applying wavelet transform to the noise reduction in interferometric SAR images is promising. Furthermore, wavelet-transform-denoise is also the adaptive algorithm based on the observation data.
-
Key words:
- synthetic aperture radar /
- noise reduction /
- noise suppression /
- wavelet /
- wavelet frame
-
1. Goldstein R M, Zebker H A, Werner C L. Satellite radar interferometry:two-dimensional phase unwrapping. Radio Science, 1988, 23(4):713~720 2. Lin Q, Vesecky J F, Zebker H A. New approach in interferometric SAR data processing. IEEE transaction on Geoscience and Remote Sensing,1992,30(3):560~567 3. Cohen A I, Daubechies I, Feauveau J C. Biorthogonal bases of compactly supported wavelets. Commun Pure Appl Math, 1992, 45(5):485~560 4. Strickland R N, Hahn H I. Wavelet transform methods for object detection and recovery. IEEE Trans on Image Proc, 1997, 6(5):724~735 5. Daubechies I. Ten lectures on wavelets. Philadelphia:SIAM, 1992 6. Mallet S,Hwang W L. Singularity detection and processing with wavelets. IEEE Trans on Inform Theory, 1992, 38(2):617~643 7. 胡庆东,毛士艺. 干涉SAR图像水平地形效应的消除及降噪方法. 北京航空航天大学学报,1999,25(4):402~405
点击查看大图
计量
- 文章访问数: 2801
- HTML全文浏览量: 275
- PDF下载量: 1100
- 被引次数: 0