摘要
根据不同特性的小波包基有效表示图像的侧重点不同,提出一种基于多个小波包基分析的图像融合去噪方法。即利用不同特性的小波包基对含噪图像进行分解,然后对软阈值处理后的系数按照一定的融合规则进行融合处理,得到一组新的融合后的子图像,最后利用小波包逆变换重构得到融合后的新图像。实验表明,本算法去噪后的图像在峰值信噪比及主观视觉效果两方面均得到了明显改善。
In view of different wavelet-bases packet effectively laying different importance on the image,a de-nosing meth-od based on multiple wavelet-bases packets is proposed in this paper using fusion technology.Different wavelet-bases packets are used to decompose the image.The coefficients after soft threshold are processed according to certain fusion rules to get a group of new sub-images.Finally the inverse transformation of wavelet packet is reconstructed to obtain the new image after fusion.Experiments show that the de-noised image by this algorithm has been improved obviously in both peak signal-to-noise （PSNR） and subjective visual effect.
出处
《长春理工大学学报（自然科学版）》
2009年第4期677-679,共3页
Journal of Changchun University of Science and Technology(Natural Science Edition)
关键词
小波包变换
图像融合
图像去噪
wavelet packet transform
image fusion
image de-nosing
