基于BPANN噪声检测的反距离加权法滤除椒盐噪声

Removal of salt and pepper noise by inverse distance weighted based on BPANN noise detection

针对传统方法滤除高密度椒盐噪声出现模糊和图像细节丢失的问题,提出基于BP神经网络噪声检测的反距离加权插值法(IDWF)滤除椒盐噪声。该算法使用有监督学习的BP神经网络检测出被椒盐噪声污染的像素点并标记;再使用反距离加权插值法对标记后的噪声图像进行重建。实验结果表明,该算法优于传统的滤波方法,修复后的图像能够保留更多的细节、拥有更高的峰值信噪比和结构相似性指数,特别是对高密度噪声图像的修复有很好的效果。

In order to solve the problem that used traditional methods to remove high density salt and pepper noise could lead to fuzzy and lose texture information,this paper proposed inverse distance weighted interpolation method(IDWF)based on check noise by BP neural network to removal salt and pepper noise.Firstly,this algorithm using supervised learning capability of back-propagation neural network to detect and mark the pixels that were corrupted by salt and paper noise.Then using inverse distance weighted interpolation method to reconstruct the marked noise image.The experimental results show that the algorithm is superior to the traditional filter methods.The restored image retains more detail features and has higher peak signal to noise ratio and structural similarity index.Performance is exceptionally good even for high density noised images.