结合双边滤波和多帧均值滤波的图像降噪

Image denoising combining bilateral and mutliframe mean filtering

针对单幅图像空间降噪中存在的问题,提出一种利用同一场景的多帧图像进行噪声抑制的自适应空域滤波方法。首先对每一帧用双边滤波器进行滤波,以在平滑噪声的同时,保护图像边缘;然后对每一像素进行帧间平均值滤波,以利用多幅图像的信息进一步滤除噪声,并减轻双边滤波后的卡通效果。提出的算法在两组多帧图像上进行了测试。实验显示,与单帧双边滤波和简单的多图像平均方法相比,该方法的降噪性能有所提高。

To deal with some problem of existing spatial denoising methods,a adaptive spatial filtering method for denoising of multi-fame images is presented.Firstly each frame is processed by bilateral filters to suppressive noise and to preserve the edge simultaneity.Then each pixel of image is filtered by meaning the same pixels of multi-frame to denoise fartherly and alleviate the cartoon-like impacts.Proposed method is tested on two image sets.The results show that it improves the performance of denoising compared with methods such as simple multi-frame meaning and single-frame bilateral filtering.