论文部分内容阅读
合成孔径雷达(SAR)海冰图像分割对全球气候研究和保证船舶航行安全具有重要意义。现有的基于区域的马尔可夫随机场(MRF)多极化SAR分割方法,由于受相干斑噪声影响,其区域划分不尽合理,不能有效完成分割。因此,提出一种噪声抑制的多极化SAR海冰图像分割算法,首先在极化总功率图上引入降低噪声的滤波算法,合理划分初始区域,其次考虑区域之间的差异度,从而实现多极化SAR海冰图像的准确分割。以RADARSAT-2和SIR-C获得的全极化海冰图像为实验数据进行验证,结果表明:和其他较先进算法相比,本文算法优势明显,既能高效保持图像连通性,又能增强图像的细节信息,具有更高的分割精度。
Synthetic Aperture Radar (SAR) sea ice image segmentation is of great significance to global climate research and to ensure the safety of ship navigation. Existing regional MRF multipolarization SAR segmentation methods, due to the speckle noise, the regional division is not reasonable, can not effectively complete the segmentation. Therefore, a noise suppression multi-polarization SAR sea ice image segmentation algorithm is proposed. Firstly, a noise reduction filtering algorithm is introduced on the polarization total power map to reasonably divide the initial region, and secondly, to consider the difference between the regions so as to achieve more Accurate Segmentation of Polarimetric SAR Sea Ice Image. The fully polarimetric sea ice images obtained by RADARSAT-2 and SIR-C are validated by the experimental data. The results show that the proposed algorithm has obvious advantages compared with other advanced algorithms. It can not only efficiently maintain the image connectivity but also enhance the image Details of the information, with higher segmentation accuracy.