基于结构保持MRF模型的SAR图像去斑
doi: 10.3724/SP.J.1146.2007.01645
A Novel MRF Model for Structure Preservation in SAR Image Despeckling
-
摘要: 该文提出一种用于SAR图像贝叶斯去斑的，具有结构保持特性的马尔可夫随机场模型。该模型通过引入一组权重参数建模待处理像素与其邻域点的空间相关性，可有效描述图像中的各种局部特性。准确的参数估计是模型有效运用的关键，借鉴随机松弛算法的思想，通过引入温度变量，实现了迭代过程中权重参数的自适应估计。仿真和实际SAR图像数据的实验结果，验证了所提模型及参数估计方法的有效性。Abstract: A novel MRF model called SPMRF (Structure Preserving MRF) is proposed for Bayesian despeckling of SAR images. In SPMRF every pixel within a neighborhood is associated with a weighting parameter which determines its contribution to the central pixels despeckling. And a temperature variable is introduced into SPMRF to control the weighting parameters estimating process which makes the model match with images local characteristics better. Experiments show as a prior model SPMRF can retain image structures efficiently with despeckling progressing.
-
贾承丽, 匡纲要. SAR图像去斑方法[J]. 中国图象图形学报,2005, 10(2): 135-141.[2]焦李成, 王爽, 侯彪. SAR图像理解与解译研究进展[J]. 电子学报, 2005, 33(12A): 2423-2434.[3]Argenti F, Bianchi T, and Alparone L. Multiresolution MAPdespeckling of SAR images based on locally adaptivegeneralized Gaussian pdf modeling [J].IEEE Trans. onImage Processing.2006, 15(11):3385-3398[4]Achim A, Kuruoglu E E, and Zerubia J. SAR image filteringbased on the Heavy-Tailed Rayleigh model [J].IEEE Trans.on Image Processing.2006, 15(9):2686-2693[5]Bratsolis E and Sigelle M. Fast SAR image restoration,segmentation, and detection of high-reflectance regions [J].IEEE Trans. on Geoscience and Remote Sensing.2003, 41(2):2890-2899[6]Walessa M and Datcu M. Model-based despeckling andinformation extraction from SAR images [J].IEEE Trans. onGeoscience and Remote Sensing.2000, 38(5):2258-2269[7]Achim A, Bezerianos A, and Tsakalides P. Novel Bayesianmultiscale method for speckle removal in medical ultrasoundimages [J].IEEE Trans. on Med. Imag.2001, 20(8):772-783 -
计量
- 文章访问数: 2912
- HTML全文浏览量: 73
- PDF下载量: 1100
- 被引次数: 0
下载:
