| 汤伟,周国庆,刘嫣,亢洁,王孟效.基于机器视觉的纸张表面缺陷在线诊断方法综述[J].中国造纸学报,2026,41(1):174-188 |
 二维码(扫一下试试看!) |
| 基于机器视觉的纸张表面缺陷在线诊断方法综述 |
| Machine Vision Based Online Diagnosis Methods for Web Surfacial Defects: A Review |
| 投稿时间:2025-03-20 修订日期:2025-04-16 |
| DOI:10.11981/j.issn.1000-6842.2026.01.174 |
| 中文关键词: 纸病在线诊断 机器视觉 图像预处理算法 纸病判定算法 纸病识别算法 |
| Key Words:online paper defect diagnosis machine vision image preprocessing algorithms paper defect determination algorithms paper defect recognition algorithms |
| 基金项目:国家自然科学基金(62073206);西安市科技计划项目(2020KJRC0146)。 |
| 作者 | 单位 | 邮编 | | 汤伟* | 1陕西科技大学电气与控制工程学院,陕西西安,710021 | 710021 | | 周国庆* | 1陕西科技大学电气与控制工程学院,陕西西安,710021 | 710021 | | 刘嫣 | 1陕西科技大学电气与控制工程学院,陕西西安,710021 | 710021 | | 亢洁 | 1陕西科技大学电气与控制工程学院,陕西西安,710021 | 710021 | | 王孟效 | 2陕西西微测控工程有限公司,陕西咸阳,712081 | 712081 |
|
| 摘要点击次数: 199 |
| 全文下载次数: 150 |
| 中文摘要: |
| 本文围绕特种纸张表面缺陷的在线诊断方法,阐述了基于机器视觉技术的纸病在线诊断的基本原理和一般流程,对其中关键技术进行了详细综述,包括纸张图像数据实时采集的硬件架构及核心设备、图像数据预处理算法(主要包括特征提取算法、图像增强算法、图像分解和重构算法)、纸病判定算法(主要包括基于灰度特征的算法、基于形态特征的算法和基于深度学习的算法)、纸病识别算法(主要包括基于特征分析的算法和基于机器学习的算法)等。此外,本文还介绍了纸病在线诊断算法的性能评价指标及后处理技术,并对当前工业场景下纸病诊断技术存在的关键问题与挑战进行了分析,对未来发展趋势进行了展望。 |
| Abstract: |
| This paper focused on online diagnosis methods for surfacial defects in specialty paper. The fundamental principles and general workflow of online paper defect diagnosis based on machine vision technology were outlined. A comprehensive review of key technologies was provided, including hardware architecture and core equipment for real-time acquisition of paper image data, image data preprocessing algorithms (mainly including feature extraction, image enhancement, and image decomposition and reconstruction algorithms), paper defect determination algorithms (mainly including grayscale feature-based, morphological feature-based, and deep learning-based algorithms), and paper defect recognition algorithms (mainly including feature analysis-based and machine learning-based algorithms). Additionally, this paper introduced performance evaluation metrics and post-processing techniques for online paper defect diagnosis algorithms. Finally, it analyzed the key issues and challenges in current industrial applications of web defect diagnosis technology and provided insights into future development trends. |
| 查看全文 查看/发表评论 下载PDF阅读器 HTML |