基于激光扫描成像技术的锯材裂缝和孔洞缺陷的-华南农业大学学报.PDF

第 29 卷第4 期 华南农业大学学报 Vo1. 29 , No.4 2008 年 10 月 lournal of 50uth China Agricultural University Oct. 2008 基于激光扫描成像技术的锯材裂缝和 孔洞缺陷的机器视觉识别系统 胡传双云 虹赵俊石2 朱新波李重根l (1华南农业大学林学院，广东广州 510642;2 北京林业大学工学院，北京 100083) 摘要:作为锯材表面常见缺陷的裂缝和孔洞，其共同特点在于所在部位厚度尺寸相较于正常材而言要小.利用激 光扫描技术形成了待检测材面的轮廓信息，把激光位移传感器输出的轮廓距离信息转换成图像的灰度值而形成轮 廓图像.根据裂缝和孔洞缺陷的形状特征，在统计的基础上提取了裂缝和孔洞缺陷的4 个识别特征，并在此基础上 开发了用于裂缝和孔洞缺陷识别的8 条规则.结果表明，所开发的基于激光扫描成像技术的锯材裂缝和孔洞缺陷 的视觉识别系统不仅可以正确表征裂缝和孔洞等厚度缺陷信息，而且能够精确地定位和分类上述缺陷. 关键词:木材缺陷;锯材;图像处理;机器视觉 中图分类号:578 1. 3 文献标识码:A 文章编号: 1001-411X(2008)04-0108-05 Locating and Identifying Splits and Knots on Sawn Lumber by a Laser Scanning Technology 1 1 1 1 HU Chuan-shuang , YUN Hong , ZHAO ]un-she , ZHU Xin-bo , LI Chong-gen (1 College offorestry ,5outh China Agricultural University , Guangzhou 510642 , China; 2 College of Engineering , Beijing Forestry University , Beijing 100083 , China) Abstract: Split and hole are two common defects on sugi ,CηptomeriαjαpomcαD. Don. They have a com- mon feature in that they are associated with surface irregularities. A laser scanning system was developed to detect splits and holes based on 出eir thickness , which correlates spatially with the profile information. Displacements measured by the laser sensor were converted to pixel values to generate displacement pro- file image. Both splits and holes manifested well in the image. A dedicated image-processing program written in Visual Basic was developed. The defect regions