基于梯度金字塔图像融合的研究

根据人体视觉对梯度的敏感性特点,采用梯度金字塔融合方法对图像进行具有方向性的多尺度、多分辨率分解,然后采用基于区域信息特征的融合规则对图像进行融合,最后进行梯度金字塔分解的逆变换得到融合图像,并对融合结果作出评价。实验结果表明,该方法对多焦点和多光谱图像融合都十分有效。

Study of Image Fusion Based on Grad Pyramid Algorithm

A Grad Pyramid Fusion algorithm based on the characteristic of the human vision is presented.The algorithm is used to decompose images in the way of multi resolution and multi scale with direction,then the decomposed images with the rule of the region character are fused,and finally obtains the fused image is finally obtained with the inverse Grad Pyramid transform.The results are evaluated and indicate that the algorithm is effective to process the multi-spectral and multi-focal images.