一种改进的多聚焦图像融合算法

An Improved Multi-focus Image Fusion Method

针对小波变换在进行多聚焦图像融合时存在的边缘失真问题,提出了一种基于双正交小波变换的图像融合算法。对图像分解后的高频分量采用一种基于区域的融合策略;对分解后的低频分量通过度量图像区域质量来选择该区域中心像素从而确定融合图像的低频分量,最后进行双正交小波反变换得到融合图像。通过对多聚焦图像的仿真实验表明,此算法得到的融合图像在清晰度和对比度方面都得到了较大提高,是一种有效的图像融合算法。

Aiming at the ring artifacts of wavelet transform in fusing multi-focus image, an image fusion method based on biorthogonal wavelet transform is proposed. The image is decomposed into lowfrequency subimage and high-frequency subimage. A region-based method is applied to high-frequency subimage to get the fused high-frequency subimage. The quality of the regions within the low-frequency subimage is measured to determine the center pixel of the fused low-frequency subimage. The inverse biorthogonal wavelet transform is performed to get the fused image at last. The clarity and contrast of the fused image are improved remarkably by this method. The algorithm is effective on fusing multi-focus image.