应用改进的弹簧质点模型进行图像滤波的算法

An Image Filtering Algorithm Using Improved Spring-Mass Model

为了克服单一使用中值滤波方法去除脉冲噪声会造成图像细节信息丢失的缺陷,提出一种基于弹簧质点模型检测的迭代中值滤波算法.首先将被检测点作为中心点,其周围8个方向的像素点对该中心点的拉力组成一个平面内的弹簧质点模型,根据弹簧质点模型的稳定条件,即平面汇交力系的平衡原理来检测像素点是否为噪声点;然后通过迭代方法,只用信号点来修改噪声点的像素值.实验结果表明,与传统的滤波算法相比,文中算法可以更有效地去除图像中的脉冲噪声并且保留原图像的细节.

There could be some information loss when the traditional median filter alone is applied to remove the impulse noise in images. To overcome this drawback, a novel iterative median filtering algorithm based on the improved spring-mass model is proposed. Firstly, the detected point is considered as a center such that the pulling forces from the eight neighboring pixels form a spring-mass model in plane. The stationary condition of the spring-mass model is used to determine whether the detected point is a noisy point. Secondly, by using iterative method, we use only signal points to modify the pixel values of the noisy points. The experimental results show that the proposed algorithm is effective in noise suppression and detail preservation compared with conventional methods.