小波变换在干涉纹图噪声识别中的应用研究

Research on Application of Wavelet Transform in Interferogram Noise Recognition

为了解决InSAR干涉纹图在数据获取和处理中易受各种噪声干扰致使图像质量变差的问题,本文从干涉纹图的生成以及相位统计特征出发,分析了干涉纹图的噪声来源,研究了基于小波域的干涉纹图噪声识别方法。以两幅不含噪声的图像作为研究对象,对两幅图像分别施加加性噪声和乘性噪声,分析其小波系数的影响值以及小波分解后各个低频系数的能量比,结果表明:不同类型噪声模型对小波系数的影响不同,加性噪声主要影响图像的高频部分,而低频部分小波系数的能量变化较小;乘性噪声主要影响图像的低频部分,而高频部分小波系数的能量变化较小。最后得出了一种根据小波系数的能量变化识别干涉纹图中噪声类型的方法。

In order to solve the problem that the image quality of InSAR interferogram is easily affected by various noise interferogram in data acquisition and processing,the noise source of interferogram is analyzed based on the generation of interferogram and phase statistical characteristics.Interferogram noise recognition method in wavelet domain.Taking two images without noise as the research object,additive noise and multiplicative noise are applied to the two images respectively,and the influence value of the wavelet coefficients and the energy ratio of each low-frequency coefficient after wavelet decomposition are analyzed.The results show that different types of noise models are shown.The influence of wavelet coefficients is different.Additive noise mainly affects the high frequency part of the image,while the energy variation of the wavelet coefficient of the low frequency part is small.Multiplicative noise mainly affects the low frequency part of the image,while the energy variation of the high frequency part wavelet coefficient is small.The research results provide a way to identify the type of noise in the interferogram based on the energy variation of the wavelet coefficients.