欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
合成孔径雷达（Synthetic Aperture Radar,简称SAR）图像受到斑点噪声影响以及成像条件的变化,使得同一场景的两幅SAR图像之间存在很大差异,利用基于边缘特征或灰度信息进行SAR图像配准,难以达到预期效果.基于此,本文提出一种基于分割区域的SAR图像配准方法,该方法首先利用主动轮廓方法对去噪后的SAR图像进行分割,得到图像分割区域;然后提取分割区域的特征;利用区域匹配度函数对图像进行粗匹配;最后利用互信息优化粗匹配下的变换参数,实现由粗到细的图像配准.实验结果验证了算法的有效性.
英文摘要：
Due to the speckle imaging and the impact of the different conditions,the synthetic aperture radar（SAR） images may differ considerably under the same scene.However,using the method of the gray or edge feature,the results of SAR image registration may not be ameliorated.Therefore,we propose a method based on the segmentation-derived region to register the SAR image.After denoising the SAR images,the segmentation regions can firstly be obtained by using the active contour image segmentation.Moreover,we extract the feature of segmentation-derived region,using the function of regional the matching degree to carry out its rough match.Finally,by using mutual information,we optimize transformation parameters such that a coarse to fine image automatic registration can be implemented.The experimental results indicate that the algorithm is effective.
同期刊论文项目
同项目期刊论文
期刊信息
位置：