论文部分内容阅读
将E-lee、E-Kuan、GammaMap、wiener等经典滤波算法和双正交小波变换相结合,提出了基于双正交小波变换域的局部统计特性SAR图像滤波方法,同时提出了一个运算量少,且是归一化的对数变换,它将乘性的Speckle噪声转为加性噪声。在小波域内建立了局部统计特性SAR图像滤波算法,使用多分辨率的手段,因为在每个方向上的小波系数都具有相同的特征,可以很好地处理图像的一些特性,使得图像边缘被模糊的相对少些。实验结果表明,此方法比经典算法的效果要好。
Combining the classical filtering algorithms such as E-lee, E-Kuan, GammaMap and wiener with biorthogonal wavelet transform, a local statistical SAR image filtering method based on biorthogonal wavelet transform domain is proposed. At the same time, , And is a normalized logarithmic transform that converts the multiplicative Speckle noise to additive noise. In the wavelet domain, a local statistical SAR image filtering algorithm is established, which uses multi-resolution method. Because the wavelet coefficients in each direction have the same characteristics, some features of the image can be well handled and the edge of the image is blurred Relatively less. Experimental results show that this method is better than the classical algorithm.