强脉冲噪声图像恢复及评价方法的研究

Investigation on Restoration and Assessment Method of Digital Image with Strong Impulse Noise

为了解决含有强脉冲噪声的数字图像在恢复及其客观质量评价中存在的问题,利用空域滤波器对图像具有的消除噪声作用,在研究强脉冲噪声数据的表示及其恢复方法的基础上,提出了一种基于点处理理论的强脉冲噪声灰度图像恢复算法。该算法由给出的恢复前后的误差公式定义表示恢复的客观质量评价方法的去噪度参数,将基于RGB模型的彩色图像可分解为R、G、B三个颜色分量,给出了将灰度图像恢复算法直接应用于各个分量的改进方法。验证实验结果表明,所提出的算法不仅能够恢复含有强脉冲噪声的灰度图像,也适用于基于RGB模型的彩色图像,且其无需大量的排序操作,具有算法执行速度快、恢复效果稳定等优点,所定义的去噪度参数能较准确地给出恢复质量的客观评价。

In order to solve the problems existed in restoration and objective quality assessment method of digital image with strong pulse noise ,employing the theory of a certain extent of noise elimination for the spatial filters, a gray image restoration algorithm of strong impulse noise based on pixel processing theory is presented on the basis of investigations on the strong pulse noise data representation and restoration method. It represents the denoising degree parameter of the restoration quality assessment method has been defined by the given formula of the error signals and decomposes the color image into the three color components ofR , G, B respectively with RGB model, and the modified method has been proposed which makes gray image restoration algorithm used in components directly. The experimental results show that the proposed algorithm can not only restore gray image with strong impulse noise,but also apply to color image based on RGB model without a large quality of the sort operations, which has advantages of fast execution and stable restoration effect, and that the definition of denoising parameter can accurately supply the objective assessment of restoration quality.