摘要
对于图像处理领域中常见的椒盐噪声来说,比较经典的去噪算法是中值滤波去噪算法.但是,中值滤波算法是直接选取一定形状的区域块,再对此区域块内图像的所有灰度值进行重排序,选取重排序后的灰度中值作为区域内新的灰度值,此方法简单但未去除噪声点对整体图像传输带来的影响.基于椒盐噪声在图片传输中的非白即黑特点,提出了一种去除噪声点影响的新算法.首先,在选定区域内去除异常噪声点,使用非噪声灰度值进行替代;然后,通过反复实验找到合理的椒盐噪声密度阈值;最后,在此阈值的情况下,进行新灰度值的排序,排序的中心值替代原始灰度排序的中心值.通过仿真验证了新算法在椒盐噪声密度比较大的场景,去噪效果较中值滤波算法更优越.
The classic image denoising algorithm is the median filter denoising algorithm for the common salt and pepper noise in the field of image processing.But the median filtering algorithm is to select an area block of a certain shape directly,and then reorder all gray values of the image in this area block.The median gray value is selected as the new gray value in the area after reordering,this method is simple however does not remove the influence of noise points on the overall image transmission.A new algorithm to remove the influence of noise points is proposed,based on the non-white and black characteristics of salt and pepper noise in image transmission.Firstly,the exceptional noise points are removed in the selected area,and the non-noise grayscale value is used instead.Then,a reasonable salt and pepper noise density threshold was found through repeated experimentations.Finally,the sorting of the new grayscale values is carried out in the case of this threshold,and the center value of the sort replaces the center value of the original grayscale sort.Simulation verifies that the new algorithm has superior denoising effect than the median filtering algorithm in scenarios with relatively large salt and pepper noise density.
作者
肖莉
XIAO Li(School of Electronic and Electrical Engineering,Anhui Sanlian University,Hefei 230601,China)
出处
《高师理科学刊》
2023年第10期41-44,共4页
Journal of Science of Teachers＇College and University
基金
安徽三联学院校级科研基金项目(KJZD2022009)
安徽省教育厅自然科学重点项目(2023AH051705,2022AH051992,2022AH052001)。
作者简介
肖莉(1987-),女,安徽合肥人,助教,硕士,从事通信与信息系统研究.E-mail:393523081@qq.com。
