摘要
图像增强过程中,如何在去除噪声的同时尽可能保留图像边缘细节非常重要.提出了一种基于极值的自适应均值滤波算法,该算法根据图像中某点的灰度值是否为邻域灰度极值将全部像素分为可疑噪声与信号两类,然后对可疑噪声点采用包括四个一维窗口和一个二维窗口在内的不同方向的五个子窗口分别计算均值,按照各个子窗口的均方差大小,自动选择窗口进行滤波,明显降低了普通均值滤波算法造成的模糊程度,使被误判的边缘像素点得到最大限度的保护.实验证明,该算法能在去除噪声的同时较好地保留边缘等细节信息,降低了图像处理后的模糊化程度,优于经典的邻域平均算法.
It is very important for the algorithm to reserve fringe detail of image while filtering noise in image enhancement.A new adaptive mean filter algorithm based on extreme value is presented.According to whether gray value of arbitrary pixel is its neighborhood extremum,all pixels are divided into doubtful noise pixels and signal pixels.Following the average variance of the five multi-scale and multidirectional windows including four one dimensional windows and one two-dimension window,the gray value of noise points are selected adaptively.Test result indicates that the algorithm is superior to classical mean filter in performance of filter noise and detail preservation.
出处
《红外与激光工程》
EI
CSCD
北大核心
2006年第z4期116-120,共5页
Infrared and Laser Engineering
关键词
均值滤波
自适应
极值
Mean filter
Adaptive
Extremum