基于直方图散度和小波分解的图像配准算法

针对基于互信息的图像配准方法运行时间长、抗噪声差的问题,提出了一种基于新的相似性测度的图像配准算法,在分析两幅图像的联合直方图点集分布情况的基础上,定义了直方图点集的散度公式,并将其作为相似性测度.为加速参数的搜索过程,配准是在小波域内进行的,并使用遗传算法与Powell算法相结合的方法来优化参数.实验证明,相对于基于互信息的图像配准算法,本算法参数优化方法选择可以更灵活,时间消耗更少,噪声鲁棒性更优.

Image registration algorithm based on histogram divergence and wavelet decomposition

The image registration algorithm based on mutual information has long runtime and bad antinoise performance.To solve this problem,an image registration algorithm based on a novel similarity measure was proposed.Histogram divergence formula was defined and used as the similarity metric by analyzing the point set distribution of joint histogram.To speed up searching the registration parameters,all were done in the wavelet field and a hybrid algorithm based on genetic algorithm and Powell's method was used to optimize this parameters.Experiments show that this algorithm can apply wider optimization methods and save more time and have better antinoise performance compared with the algorithm based on mutual information.