一种尺度自适应调整的高斯滤波器设计方法

An adaptive Gaussian filter with scale adjustable

固定尺度参数的经典高斯滤波器往往在滤除噪声的同时造成图像的模糊,为此设计了一种尺度可变的自适应高斯滤波器。通过计算滤波器各滤波窗口内像素点灰度值的均值,以及该均值与当前像素点灰度值的差值,确定滤波窗口的平滑度,并将其作为高斯函数尺度的大小,从而实现滤波器尺度参数的自适应调整。实验结果表明,该滤波器在有效滤除噪声的同时能够尽可能的保持图像原有信息不受影响,且计算简单,适于工程应用。

Using classical Gaussian filter with fix scale-parameter for noise suppression, often the noise is smoothed out, while the signal is also distorted. Therefore an adaptive Gaussian filter with the scale adjustable was designed. By computing the average of all intensity values in current filter window, and the difference between the average and the intensity value of the cur- rent pixel, the smooth degree of current filter window was obtained. And then by regarding the smooth degree as the scale size of Gaussian function, the scale parameter of the filter was adjusted automatically. Experiment results proved that this filter can remove noises effectivdy with the original signals of images preserved as much as possible. It is simple and adapt to engineering applications.