摘要
研究了基于最大互信息图像配准的理论和实现算法，提出了将小波分解与之相结合的算法，证明了该算法优于基于最大互信息的图像配准。通过实验数据比较可以得出，提出的基于小波分解和最大互信息的图像配准的算法可以有效地提高配准的速度和具有较好的配准效果。
This paper studies the theory and implementation algorithm of image registrat/on based on the maximum mutual information. And on that basis, the paper presents an idea of algorithm improvement on the combination of the wavelet decomposition with the maximum mutual information. The processing speed of matching can be greatly improved by using approximate frequency of wavelet decomposition and the paper also proves by the theory that the improved algorithm is superior to maximum mutual information image registration. By comparing the different experimental dates, the image registration algorithm based on wavelet decomposition and maximum mutual information proposed in this paper increases the matching speed effectively and shows better matching results.
出处
《陕西理工学院学报（自然科学版）》
2010年第3期22-26,共5页
Journal of Shananxi University of Technology:Natural Science Edition
关键词
最大互信息
小波分解
图像配准
maximum mutual information
wavelet decomposition
image registration
