基于第二代Bandelet和区域特性的多聚焦图像融合算法

Multi-focus image fusion algorithm based on Bandelet and region feature

提出了一种能充分利用第二代Bandelet变换自适应捕获图像边缘的能力,并基于区域特性进行融合的多聚焦图像融合新算法。首先将源图像分解至Bandelet变换域,然后结合区域特性进行融合处理:对几何流,采用绝对值最大的融合规则;对Bandelet系数矩阵采用区域方差的融合规则,最后通过Bandelet逆变换得到融合图像。实验结果表明,所提出的新算法能够更好地提取参与融合的原始图像的基本特征进行融合处理,融合效果在主观视觉效果和客观性能指标两方面都较经典的拉普拉斯金字塔算法和小波变换算法更具优势,尤其是对纹理及边缘信息明显的源图像,融合结果优势更加明显。

A multi-focus image fusion algorithm based on Bandelet transform and region statistics was developed,which could fully utilize the second generation Bandelet's advantages of the geometrical regularity of image structure and represent sharp image transitions such as edges efficiently in image fusion source images were firstly decomposed by the Bandelet transform,for reconstructing the fused image,the maximum rule was used to select source images' geometric flow and regional variance was used to the Bandelet coefficients.Finally the fused image was reconstructed by performing the inverse bandelet transform.The experimental results indicate that the Bandelet-based fusion algorithm represents the edge and detailed information well and outperforms the wavelet-based and Laplacian pyramid-based fusion algorithms,especially when the abundant texture and edges are contained in the source images.