一种改进均值的自适应中值滤波算法
|
摘要
针对自适应中值滤波算法在滤除高浓度椒盐噪声和保留图像边缘细节中的不足,提出了一种改进均值的自适应中值滤波(IMAMF)算法。该算法采用扩充图像边界的方式,使得原图像的边界点能在自适应的滤波窗口下参与噪声检测和滤波处理,并在检测噪声和信号时,增加了噪声阈值判定,将存在噪声的像素点用修正后的均值滤波器值输出,信号点则用原始灰度值输出。为了验证算法的可行性,采用了5种不同的算法进行仿真对比分析,并从主观角度和客观指标上进行效果评价。试验结果表明:该算法能有效滤除浓度为10%～90%范围内的椒盐噪声,且图像细节和边缘信息得到了更好的保留,滤波性能明显优于其他算法。
To deal with the deficiency in terms of de-noising and preserving image details of the adaptive median filter,an improved mean-valued method of adaptive median filtering( IMAMF) is proposed. The algorithm adopts the way of extending edge of the image,so that all the pixels of the original image can be detected by noise detection factor and filtering under the adaptive window. When noise and signal are detected,the noise threshold is increased. If the pixels with noise are corrected,the new average filter value is output.Otherwise,the signal point is output with the original gray value. In order to verify the feasibility of the algorithm,five different algorithms are used for simulation analysis. From the subjective and objective angles of different algorithms for comparative analysis,the experimental results show that the algorithm can not only effectively filter out salt and pepper noise in the range of 10% to 90%,but also can better preserve the details and edge information of the original image. Obviously,the new method has better properties.
To deal with the deficiency in terms of de-noising and preserving image details of the adaptive median filter,an improved mean-valued method of adaptive median filtering( IMAMF) is proposed. The algorithm adopts the way of extending edge of the image,so that all the pixels of the original image can be detected by noise detection factor and filtering under the adaptive window. When noise and signal are detected,the noise threshold is increased. If the pixels with noise are corrected,the new average filter value is output.Otherwise,the signal point is output with the original gray value. In order to verify the feasibility of the algorithm,five different algorithms are used for simulation analysis. From the subjective and objective angles of different algorithms for comparative analysis,the experimental results show that the algorithm can not only effectively filter out salt and pepper noise in the range of 10% to 90%,but also can better preserve the details and edge information of the original image. Obviously,the new method has better properties.
引文
[1]苟中魁,少军,忠富,等.一种基于极值的自适应中值滤波算法[J].红外与激光工程,2005,34(1):98-101.
[2]郭红伟.数字图像中椒盐噪声的滤波算法研究[D].昆明:云南大学,2010.
[3]雷浩鹏.数字图像去噪算法研究及应用[D].长沙:长沙理工大学,2010.
[4]王琪洁,朱建军,丁晓利,等.基于自适应中心加权中值和Goldstein的In SAR干涉图强噪声滤波方法[J].测绘通报,2009(8):13-16.
[5]兰霞,刘欣鑫,沈焕锋,等.一种消除高密度椒盐噪声的迭代中值滤波算法[J].武汉大学学报(信息科学版),2017,42(12):1731-1737.
[6]钟涛,张建国,左俊彦.一种改进的中值滤波算法及其应用[J].云南大学学报(自然科学版),2015,37(4):505-510.
[7]张恒,雷志辉,丁晓华.一种改进的中值滤波算法[J].中国图象图形学报,2004,9(4):26-29.
[8]WANG Z,HANG D.Progressive switching median filter for the removal of impulse noise from highly corrupted images[J].IEEE Transactions on Circuits and Systems II:Analog and Digital Signal Processing,1999,46(1):78-80.
[9]张思杰,喻洁,徐鹏,等.加密图像的开关中值滤波去噪算法[J].重庆大学学报,2014,37(2):113-116.
[10]秦鹏,丁润涛.一种基于排序阈值的开关中值滤波方法[J].中国图象图形学报,2004,9(4):30-34.
[11]SUN T,NEUVO Y.Detail-preserving median-based filters in image processing[J].Pattern Recognition Letters,1994,15(4):341-347.
[12]孙树亮,王守觉.一种基于改进的极值中值滤波算法[J].计算机科学,2009,36(6):165-166.
[13]黄存令,段锦,祝勇,等.一种改进的极值中值滤波算法[J].长春理工大学学报(自然科学版),2013,36(3):141-143.
[14]HWANG H,HADDAD R.Adaptive median filters:new algorithms and results[J].IEEE Transactions on Image Processing,1995,4(4):499-502.
[15]王松林,蒋峥.改进的自适应加权中值滤波算法[J].传感器与微系统,2016,35(11):128-131.
[16]ZHANG X,XU B.Adaptive median filtering for image processing[J].Journal of Computer Design&Computer Graphics,2005,17(2):295-299.
[17]VIEIRA M N,SANSAO J P H,YEHIA H C.Measurement of signal-to-noise ratio in dysphonic voices by image processing of spectrograms[J].Speech Communication,2014,61(6):17-32.
[18]张培轩.一种基于椒盐噪声的自适应加权均值滤波算法[D].上海:华东师范大学,2015.
[2]郭红伟.数字图像中椒盐噪声的滤波算法研究[D].昆明:云南大学,2010.
[3]雷浩鹏.数字图像去噪算法研究及应用[D].长沙:长沙理工大学,2010.
[4]王琪洁,朱建军,丁晓利,等.基于自适应中心加权中值和Goldstein的In SAR干涉图强噪声滤波方法[J].测绘通报,2009(8):13-16.
[5]兰霞,刘欣鑫,沈焕锋,等.一种消除高密度椒盐噪声的迭代中值滤波算法[J].武汉大学学报(信息科学版),2017,42(12):1731-1737.
[6]钟涛,张建国,左俊彦.一种改进的中值滤波算法及其应用[J].云南大学学报(自然科学版),2015,37(4):505-510.
[7]张恒,雷志辉,丁晓华.一种改进的中值滤波算法[J].中国图象图形学报,2004,9(4):26-29.
[8]WANG Z,HANG D.Progressive switching median filter for the removal of impulse noise from highly corrupted images[J].IEEE Transactions on Circuits and Systems II:Analog and Digital Signal Processing,1999,46(1):78-80.
[9]张思杰,喻洁,徐鹏,等.加密图像的开关中值滤波去噪算法[J].重庆大学学报,2014,37(2):113-116.
[10]秦鹏,丁润涛.一种基于排序阈值的开关中值滤波方法[J].中国图象图形学报,2004,9(4):30-34.
[11]SUN T,NEUVO Y.Detail-preserving median-based filters in image processing[J].Pattern Recognition Letters,1994,15(4):341-347.
[12]孙树亮,王守觉.一种基于改进的极值中值滤波算法[J].计算机科学,2009,36(6):165-166.
[13]黄存令,段锦,祝勇,等.一种改进的极值中值滤波算法[J].长春理工大学学报(自然科学版),2013,36(3):141-143.
[14]HWANG H,HADDAD R.Adaptive median filters:new algorithms and results[J].IEEE Transactions on Image Processing,1995,4(4):499-502.
[15]王松林,蒋峥.改进的自适应加权中值滤波算法[J].传感器与微系统,2016,35(11):128-131.
[16]ZHANG X,XU B.Adaptive median filtering for image processing[J].Journal of Computer Design&Computer Graphics,2005,17(2):295-299.
[17]VIEIRA M N,SANSAO J P H,YEHIA H C.Measurement of signal-to-noise ratio in dysphonic voices by image processing of spectrograms[J].Speech Communication,2014,61(6):17-32.
[18]张培轩.一种基于椒盐噪声的自适应加权均值滤波算法[D].上海:华东师范大学,2015.