一种基于Curvelet变换多传感器图像融合算法

Fusion of Multisensor Images Based on the Curvelet Transform

综合分析了Curvelet变换的特性,并提出了一种基于Curvelet变换的多传感器图像融合算法。首先采用Curvelet变换将源图像分解到不同尺度、方向频带范围内,然后采用不同的融合规则得到融合后图像的Curvelet变换系数,最后再进行Curvelet逆变换得到融合图像。采用多组具有不同特征的源图像进行了融合实验,并对融合图像进行了主客观评价。实验结果表明,相比于传统的基于小波变换的图像融合算法,该算法能够有效避免“人为”效应或高频噪声的引入,得到具有更好视觉效果和更优量化指标的融合图像。

The characteristics of the Curvelet transform are studied, and a novel algorithm to fuse multi-sensor images based on the Curvelet transform is proposed in this paper. Firstly, the curvelet transform is used to perform a multi-scale and multi-orientation decomposition of each image. And then,the Curvelet coefficients for the fused image can be obtained by means of different fusion rules. Finally,the fused image is reconstructed by the inverse Curvelet transform. The proposed method is successfully used to merge several sets of multisensor images with different modalities. The experimental results indicate that the proposed approach can avoid the introduction of the artifacts and the high frequency noise and can significantly outperform the traditional wavelet-transform-based image fusion method in terms of both visual quality and objective evaluation criteria.