改进中值滤波算法在晶棒图像预处理中的应用
Application of improved median filtering algorithm in pre-processing of crystal rod images
邓志诚 1黄卫东 1吴仙星2
作者信息
- 1. 福建省智能加工技术及装备重点实验室,福建 福州,350118;福建理工大学 机械与汽车工程学院,福建 福州 350118
- 2. 福建省南平智创技术服务有限公司,福建 南平 353000
- 折叠
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.
Abstract
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 me-dian filtering algorithm based on image gradients was proposed.By introducing pixel similarity and image gra-dient information,the accuracy of noise removal and the ability to protect line features were improved.A com-parison 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 me-dian 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Key words
crystal rod/median filtering/image gradient/PSNR引用本文复制引用
出版年
2024
NETL
NSTL
万方数据