摘要
人造木质板材表面的划痕、油污、缺损和花色缺失是影响产品质量的关键因素,工业自动化生产线上需要在线筛查与分拣。基于高实时性需求及人造木质板材缺陷背景的复杂性,为了能够快速准确地识别人造木质板材的表面缺陷,提出了一种基于多特征融合的人造木质板材缺陷检测算法。该算法通过对样品进行预处理并获取人造木质板材的色彩特征和纹理特征,建立相应特征库;通过待检测板材所提取特征与特征库进行对比分析,达到缺陷检测的目的。针对特征匹配容易出现误判的问题,利用代价复杂度算法对多个特征建立特征库,实现多维度特征匹配。实验结果表明,该检测方法能够准确识别板材表面缺陷,准确率可达98%,可以满足人造木质板材工业生产中对缺陷自动化识别准确性的基本要求。研究结果可以为人造木质板材在线缺陷检测提供参考。
Scratches,oil stains,defects and lack of color on the surface of artificial wood board are key factors affecting product quality.Industrial automation production lines require online screening and sorting.Based on the high real-time requirements and the complexity of the defect background of artificial wood board,an artificial wood board defect detection algorithm based on multi-feature fusion is proposed in order to quickly and accurately identify its surface defects.The algorithm preprocessed the samples and obtained the color and texture features of the artificial wood board,established the corresponding feature data base,and compared and analyzed the extracted features of the board to be tested with the feature data base to achieve the purpose of defect detection.Aiming at the problem of feature matching prone to misjudgment,the cost complexity algorithm was used to build a feature data base for multiple features to realize multi-dimensional feature matching.The experimental results show that the detection method can accurately identify the surface defects of the board,and that the accuracy rate can reach 98%,which meets the basic requirements for the accuracy of automatic defect recognition in the industrial production of artificial wood board.The research results can provide a reference for the online defect detection of artificial wood board.
作者
徐浩
夏振平
林李兴
顾敏明
XU Hao;XIA Zhenping;LIN Lixing;GU Minming(School of Physical Science and Technology,SUST,Suzhou 215009,China;School of Electronic&Information Engineering,SUST,Suzhou 215009,China)
出处
《苏州科技大学学报（自然科学版）》
CAS
2024年第1期76-84,共9页
Journal of Suzhou University of Science and Technology（Natural Science Edition）
基金
国家自然科学基金项目(62002254)
江苏省自然科学基金项目(BK20200988)。
关键词
人造木质板材
多特征融合
色彩特征
纹理特征
在线检测
artificial wood board
multi-feature fusion
color feature
texture feature
online detection
