当前位置: X-MOL 学术Eur. J. Wood Prod. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Automatic detection of pith location along norway spruce timber boards on the basis of optical scanning
European Journal of Wood and Wood Products ( IF 2.4 ) Pub Date : 2020-07-16 , DOI: 10.1007/s00107-020-01558-1
Tadios Habite , Anders Olsson , Jan Oscarsson

Knowledge of annual ring width and location of pith in relation to board cross-sections, and how these properties vary in the longitudinal direction of boards, is relevant for many purposes, such as assessment of shape mechanical properties and stability of sawn timber. Hence, the present research aims at developing a novel method and an algorithm, based on data obtained from optical surface scanning, by which the pith location along the length of sawn timber boards can be determined accurately and automatically. The first step of the method is to identify clear wood sections, free of defects along boards. Then time-frequency analysis, using the continuous wavelet transform, is applied to detect the surface annual ring width distribution of the four sides of the selected sections. Finally, the pith location is estimated by comparing annual ring width distributions on the different surfaces, and assuming that annual rings are concentric circles with the pith in the centre. The proposed algorithm was applied to a total sample of 104 Norway spruce boards. Results indicate that optical scanners and the suggested automatic method allow for accurate detection of annual ring width and location of pith along boards. For a sample of boards with the pith located within the cross-section, a mean error of 2.6 mm and 3.2 mm in the depth and thickness direction, respectively, was obtained. For a sample of boards of which 60% with pith located outside the cross-section, a mean discrepancy between automatically and manually determined pith locations of 3.9 mm and 5.8 mm in depth and thickness direction, respectively, was obtained.



中文翻译:

基于光学扫描自动检测挪威云杉木板的髓位

了解年轮宽度和髓相对于木板横截面的位置,以及这些性质在木板的纵向上如何变化,对于许多目的都是有意义的,例如评估形状机械性质和锯材的稳定性。因此,本研究的目的是基于从光学表面扫描获得的数据,开发一种新颖的方法和算法,通过该方法和算法,可以准确,自动地确定沿锯切木板长度的髓位。该方法的第一步是识别出没有木板缺陷的透明木材部分。然后,使用连续小波变换进行时频分析,以检测所选截面的四个面的表面年轮宽度分布。最后,通过比较不同表面上的年轮宽度分布,并假设年轮是同心圆,而中心位于中心,则可以估算出髓的位置。该算法被应用于104个挪威云杉板的总样本中。结果表明,光学扫描仪和建议的自动方法可精确检测年轮宽度和沿木板的髓的位置。对于髓位于横截面内的木板样品,在深度和厚度方向的平均误差分别为2.6 mm和3.2 mm。对于其中60%的髓位于横截面之外的木板样品,自动和手动确定的髓位置在深度和厚度方向分别为3.9 mm和5.8 mm,这是平均差异。

更新日期:2020-07-16
down
wechat
bug