基于计算机断层扫描技术的木材节子检测算法研究

介绍了目前国内外通过计算机断层扫描技术(CT)结合各种算法对木材中节子部分进行无损检测所取得的研究进展,概述了以灰度阈值法、滤波算法、最大似然法和神经网络法为主的算法识别特点,并对其中应用广泛的神经网络算法进行了对比分析。现有研究表明,运用该技术结合多种算法可实现对原木中节子参数特征的提取与分析,通过算法的不断改进能够提高节子检测的准确率。文中还总结了CT技术在处理节子检测方面存在的主要问题,并对未来趋势进行了展望。

Computed Tomography(CT) Technology Application in Knot Algorithm Study

The paper introduced the domestic and international research progress on non-destructive testing of wood knot by computed tomography technology combined with various algorithms, described algorithm's identifying characteristics in gray threshold method, filtering algorithm, maximum likelihood method and neural network algorithm, and comparatively analyzed the wide application of neural network algorithm. Existing research shows that the use of CT technique combined with a variety of algorithm can achieve extraction and analysis of knot parameters features, and the accuracy in knot test could be improved through the continuous improvement of the algorithm. Finally, the main problems existing in knot testing with CT technology were summarized and the future tendency was also discussed.