SAR图像去噪方法研究本科.doc

摘 要 合成孔径雷达(SAR)图像的相干斑噪声严重降低了图像的可解译度，影响了后续目标检测、分类和识别等应用。SAR图像的相干斑噪声是成像过程中出现的原理性缺陷。在过去的二十年里，国内外的诸多学者提出了大量抑制图像相干斑噪声的方法。其中大多数方法是利用一个定义好的滤波器窗口来估计相干斑图像局部噪声方差，然后在局部分别作滤波处理。结果表明这些方法在均匀的区域内能够大大削减相干斑噪声，而在非均匀区域内图像变得模糊或过于平滑而失去细节信息，降低了图像的空间分辨率，图像的解译性因此变差。理想的降噪方法是能在很好地保留边缘和细节信息的前提下抑制相干斑噪声。 本论文的研究内容： （1）首先介绍了SAR图像及其噪声的国内外研究动态，然后简述了SAR图像噪声的形成机理。 （2）重点对SAR图像斑点噪声的抑制算法进行了研究，对经典斑点噪声抑制算法(卷积滤波中的中值滤波；还有基于统计估计理论的Lee滤波算法及其增强算法、Kuan滤波算法、Frost滤波算法、小波滤波算法、均值滤波算法等)进行了原理说明。 （3）结合实例，将各种算法的处理结果进行了分析和比较，对其效果和优缺点进行了评价。 关键词：合成孔径雷达；噪声抑制；效果评价 ABSTRACT The presence of speckle noise in the synthetic aperture radar (SAR)images。severely restricts the appl ication of the coherent images．Speckle noise is generated by the coherent processing of radar signals．During the past two decades，many speckle reduction techniques have been developed for removing ospeckle．An ideal algorithm should smooth the speckle without blurred edges and fine detail．But most algorithms cannot satisfy these two demands very well． This paper the research content: (1) First introduced the SAR and the noise of the image research dynamic, then briefly introduced the formation mechanism of the SAR image noise. (2) Focus on studying the algorithm of image noise suppression of SAR spots, introduced the principle of the classic spots noise suppression algorithm ( the convolution filter median filtering, and based on the statistical estimation theory of filter algorithm and its enhancement algorithm Lee Kuan, filter algorithm, Frost filter algorithm, wavelet filter algorithm, average filtering algorithm, etc). (3) Combined with an example, all with the result of the algorithm is analyzed and compared, evaluated their effects, advantages and disadvantages. Keywords: SAR; Noise inhibition; Effect evaluation 目 录 1绪论 1 1.1 SAR图像噪声抑制研究背景及其意义 1 1.2 SAR图像纹斑噪声抑制的研究现状 2 1.3本论文的研究内容 6 2 SAR图像常用去噪方法介绍 8 2.1 SAR图像噪声产生机理 8 2.2传统的SAR图像去噪方法 9 2.2.1均值滤波 9 2.2.2中值滤波 10 2.3基于自适应滤波的SAR图像去噪方法 10 2.3.1 Lee 滤波 11 2.3.2增强 Lee滤波 11 2.3.3