基于分数阶GM(1,1)的图像去噪

Image Denoising Based on Fractional Order GM(1,1)

针对传统整数阶GM(1,1)模型无法调节阶次来改善图像去噪性能,采用分数阶GM(1,1)模型来弥补其不足,分数阶GM(1,1)模型可以精确调节累加数之间数量级来达到更好的去噪效果.先通过在经典图像中添加椒盐噪声的去噪对比实验,得出分数阶GM(1,1)模型较GM(1,1)模型有更好的视觉效果,更高的峰值信噪比和结构相似度等去噪性能,从而验证了分数阶GM(1,1)模型的优势.然后进一步将分数阶GM(1,1)模型与文献中几种去噪模型进行对比实验,同样得到分数阶GM(1,1)模型具有更好的去噪效果.

Fractional order GM(1,1)model is used to compensate for the inadequacy of traditional integer order GM(1,1)model which can not adjust the order to improve theimage denoising performance.Fractional order GM(1,1)model can accurately adjust the order of magnitude between the accumulating numbers to achieve better denoising effect.Firstly,by adding salt and pepper noise to the classical image,it is concluded that fractional order GM(1,1)model has better visual effect,higher peak signal-to-noise ratio,structural similarity and so on than the GM(1,1)model,thus verifying the advantages of fractional order GM(1,1)model.Then,fractional order GM(1,1)model is compared with several denoising models in the literature,and fractional order GM(1,1)model has better denoising effect.