An image denoising method based on robust regression is proposed. It detects the edge pixels using optimal-size adaptive window technique. This technique prevents the edge from breaking the denoising effect and utilizes the neighborhood pixels as much as possible to improve the performance of the regression. Investigation on real images contaminated by mixture Gaussian noise has demonstrated the present method has good performance and is reliable under different noise distributions.%提出一种基于稳健回归的图像去噪方法.该方法使用最优尺寸自适应窗技术来探测图像的边界像素,减少边界像素对降噪算法的影响,并且使用尽可能多的邻域信息来提高回归性能.使用该方法对加入混合高斯噪声的真实图像进行降噪实验,结果表明,该方法不仅具有出色的去噪性能,而且能够适应不同的噪声分布.
展开▼
