计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (34): 193-195.

• 图形、图像、模式识别 • 上一篇    下一篇

自适应控制迭代的随机值脉冲噪声滤波

郭红伟1,2,余 江2,骆洪军1,张自红1   

  1. 1.红河学院 工学院,云南 蒙自 661100
    2.云南大学 信息学院,昆明 650091
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-12-01 发布日期:2011-12-01

Adaptive iterative control for random values impulse noise filtering

GUO Hongwei1,2,YU Jiang2,LUO Hongjun1,ZHANG Zihong1   

  1. 1.Engineering College,Honghe University,Mengzi,Yunnan 661100,China
    2.Information College,Yunnan University,Kunming 650091,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-12-01 Published:2011-12-01

摘要: 定义了一种更接近实际噪声情况的随机值脉冲噪声模型,针对随机值脉冲噪声的滤除提出一种自适应控制迭代次数的滤波算法。算法包括噪声检测、噪声滤除、误检像素修正和迭代控制四个步骤,对固定值脉冲噪声和随机值脉冲噪声均能有效滤除。与中值滤波算法进行实验比较,在滤除随机值脉冲噪声时,该算法滤波后图像细节信息保护较好,且滤波过程无需设定迭代参数,自适应性强。

关键词: 脉冲噪声, 中值滤波, 迭代, 自适应性

Abstract: The random value impulse noise modeling is defined,it is more realistic noise conditions,an adaptive controlling iteration times filter algorithm is proposed for the random value impulse noise remove.Algorithm consists of four steps,noise detection,noise filtering,error detection pixel correction and iterative control.The algorithm can remove fixed-value noise and random-value impulse noise effectively,and can effectively filter out fixed value impulse noise and random value impulse noise.Compared with median value filtering algorithm,this algorithm can protect filtered image information better,and there is no need to set the iteration parameters in the filtering process,with strong adaptability.

Key words: impulse noise, median filter, iteration, adaptability