一种基于动态阈值的自适应中值滤波方法  

An Adaptive Median Filtering Algorithm Based on Dynamic Threshold

作　　者：池利阳[1] 师翔[1] 吴新强[1] 

机构地区：[1]桂林电子科技大学计算机科学与工程学院,桂林541004

出　　处：《现代计算机（中旬刊）》2014年第10期58-60,共3页Modern Computer

摘　　要：针对图像受到电磁干扰带来的脉冲噪声,对自适应中值滤波方法进行改进。在去除噪声的过程中为了更好地将信号和噪声进行区分开,引入噪声点的预判机制。为满足改进算法对自适应中值滤波半径和窗口的变化的适应,采用自适应动态阈值。通过与SMF、RAMF等方法的比较实验表明,该算法具有较高的峰值信噪比和边缘保持性能。To overcome the impulse noise caused by the electromagnetic interference, proposes an adaptive median filter method. In order to put the signal and noise are separated in the process of removing noise, introduces noise prediction mechanism. In order to adapt to the change of adaptive median filter radius and window, uses the adaptive dynamic threshold. The results of comparison experiments with SMF, RAMF demonstrate that the new algorithm has higher peak signal to noise ratio and edge preserving performance.

关 键 词：脉冲噪声 中值滤波 自适应 峰值信噪比 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]