论文部分内容阅读
为改进SAR图像匹配的稳健性和实时性,提出一种基于小波变换的等价图割SAR图像配准方法.该方法首先利用小波变换对图像进行分解,在低频子图像下构造等价图割,克服相干斑噪声干扰,避免NP困难,解决映射函数选取问题,从图像中分割出精确目标.其次利用尺度不变特征变换(SIFT)方法实现目标的特征匹配,降低搜索空间特征点描述,提高实时性.最后通过匹配关系找到变换参数,实现图像精确配准.实验结果表明,该方法能快速而精确地实现SAR图像配准.
In order to improve the robustness and real-time performance of SAR image matching, an image segmentation method based on wavelet transform is proposed.It first decomposes the image by using wavelet transform and constructs the equivalent image segmentation , To overcome the interference of speckle noises, to avoid NP difficulties, to solve the problem of selecting the mapping function, and to segment the exact target from the image.Secondly, the feature invariants of SIFT are used to reduce the feature points of the search space Real-time.Finally, the transformation parameters are found through the matching relationship to achieve accurate image registration.Experimental results show that this method can achieve SAR image registration quickly and accurately.