改进中值滤波算法在晶棒图像预处理中的应用  

Application of improved median filteringalgorithm in pre-processing of crystal rod images

作　　者：邓志诚 黄卫东 吴仙星 DENG Zhicheng;HUANG Weidong;WU Xianxing(Key Laboratory of Intelligent Processing Technology and Equipment in Fujian Province,Fuzhou 350118,China;School of Mechanical and Automotive Engineering,Fujian University of Technology,Fuzhou 350118,China;Fujian Nanping Zhichuang Technology Service Co.,Ltd.,Nanping 353000,China)

机构地区：[1]福建省智能加工技术及装备重点实验室,福建福州350118 [2]福建理工大学机械与汽车工程学院,福建福州350118 [3]福建省南平智创技术服务有限公司,福建南平353000

出　　处：《福建理工大学学报》2024年第3期275-279,共5页JOURNAL OF FUJILAN UNIVERSITY OF TECHNOLOGY

基　　金：福建省创新资金项目(2022C0051)。

摘　　要：传统的中值滤波算法在晶棒图像处理中无法准确区分噪声和划线特征,导致划线特征过度平滑。为解决该问题,提出了一种基于图像梯度的中值滤波算法。通过引入像素相似度以及图像梯度信息来提高噪声去除的准确性和划线特征保护能力。比较处理后图像的峰值信噪比和划线特征边缘梯度强度,结果显示,经改进算法处理后,图像的划线特征边缘强度和峰值信噪比均高于标准中值滤波和快速加权中值滤波算法,在晶棒图像预处理中性能更优。Traditional median filtering algorithms cannot accurately distinguish between noise and line features in crystal rod image processing,resulting in excessive smoothing of line features.To address this issue,a median filtering algorithm based on image gradients was proposed.By introducing pixel similarity and image gradient information,the accuracy of noise removal and the ability to protect line features were improved.A comparison was conducted of the peak signal-to-noise ratio and edge gradient intensity of the processed images,and results show that after the improved algorithm processing,the edge intensity and peak signal-to-noise ratio of the line features in the images were higher than those of the standard median filtering and fast weighted median filtering algorithms,and had better performance in the preprocessing of crystal rod images.

关 键 词：晶棒 中值滤波 图像梯度 峰值信噪比 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]