摘要
提出了一种快速易行的时间序列图像自动配准算法:首先对两幅图像分别进行离散小波变换,然后对分解后的两幅低频图像计算联合直方图,最后采用改进的联合直方图区域计数方法来设计目标函数并进行配准参数优化。实验结果与对比分析表明:在基本不影响图像配准精度的情况下,算法大大加快了图像配准速度。
Serial images refer to an aggregate of images of the same scene captured in serial time, This paper proposes a new method to solve the problem of serial images registration. The source images are decomposed by the wavelet transform. The lowest frequency bands of the source images are used to estimate the registration parameters. The traditional joint histogram based registration algorithm is improved by applying new region-counting method. Experiments show that the proposed algorithm can increase the soeed of registration.
出处
《计算机工程》
CAS
CSCD
北大核心
2006年第11期228-230,共3页
Computer Engineering
基金
国家自然科学基金资助项目(60473103
60473127)
