机电工程技术2025，Vol.54Issue(6)：11-15,5.DOI:10.3969/j.issn.1009-9492.2025.06.003
基于改进中值滤波算法的拉索表面缺陷图像降噪方法研究
Research on Denoising Method of Cable Surface Defect Image Based on Improved Fast Median Filter Algorithm
摘要
Abstract
The traditional median filtering algorithm has the problem that the noise reduction effect is not good and the time is long when removing the image noise of cable surface defects,which leads to the failure of accurate detection of cable surface defects.Aiming at this problem,an improved median filtering algorithm is presented.First,the extreme noise points in the image are detected,and the mean values of the surrounding pixels are replaced.Secondly,the filtering window is set,the difference between the mean value of the filtering window and the absolute value of each pixel are calculated,and the median value of the window is found through the absolute difference.Finally,the filter window is moved to compare the pixel value entering the window and the pixel value removing the window.If the gray value of the pixel is the same,the median value remains unchanged.If different,determine the new window median.By improving the median filter,median filter and mean filter,the image of paint layer shedding defect on the cable surface of Lanzhou Yintan yellow river bridge is denoised.The results show that,compared with the other two noise reduction algorithms,the PSNR value of the processed image is increased by 3.47～5.9,and the noise reduction time is shortened by 0.41～0.48 s.The proposed method can effectively remove the noise in the image,retain more detail information of the cable image,and the processing time is shorter.关键词
改进中值滤波/图像降噪/拉索表面缺陷Key words
improved median filtering/image noise reduction/cable surface defect分类
土木建筑引用本文复制引用
张跃,殷红..基于改进中值滤波算法的拉索表面缺陷图像降噪方法研究[J].机电工程技术,2025,54(6):11-15,5.基金项目
甘肃省自然科学基金(23JRRA907) （23JRRA907）
