摘要
针对Curvelet分解的不同频率域,分别讨论了低频系数和高频系数的选择原则。在选择低频系数时,采用了基于边缘的方案。在选择高频系数时,充分利用Curvelet变换具有方向性的优点,提出了Curvelet域区域边缘的概念,并给出了基于区域边缘的系数选择方案。实验结果表明:所给出的融合算法能够很好地保留多幅源图像中的边缘信息,得到多个目标都清晰的图像。
According to the different frequency bands decomposed by Curvelet transform, the selection principles of low frequency coefficients and high frequency coefficients are discussed respectively. In choosing the low frequency coefficient, image edge concept is employed. In choosing the high frequency coefficient,taking the advantage of the directional characteristic of the Curvelet transform, a concept of area edge in the Curvelet domain is proposed and a selection principle based on area edge is suggested. The experimental resuhs show that the proposed algorithm can extract all useful information from the original images and make all targets in the fused images very clear.
出处
《信息化研究》
2009年第3期16-19,共4页
INFORMATIZATION RESEARCH
