Calibration of SilviScan data of Cryptomeria japonica wood concerning density and microfibril angles with NIR hyperspectral imaging with high spatial resolution
Calibration of SilviScan data of Cryptomeria japonica wood concerning density and microfibril angles with NIR hyperspectral imaging with high spatial resolution
复制标题
利用高空间分辨率近红外高光谱成像对日本柳杉木材密度和微纤维角度的 SilviScan 数据进行校准
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
S. Tsuchikawa
中科院分区：
文献类型：
--
作者：
Te Ma;T. Inagaki;S. Tsuchikawa
摘要木材密度和微纤丝角与木材劲度、胀缩性及其各向异性密切相关。了解这些数据的空间分布对于实木应用至关重要。在这项研究中，近红外（NIR）高光谱成像已被校准的木材密度和MFA的评价在一个有效的方式。简而言之，五个木材样品收集两个不同的柳杉树的正常木材（NW）和压缩木材（CW）部分进行了分析。采用偏最小二乘（PLS）回归分析确定SilviScan和NIR光谱的X射线光密度测量数据之间的关系，并采用留一法（leave-one-out）进行交叉验证。通过近红外技术预测的密度与X射线数据之间的验证决定系数（r2）为0.83，交叉验证的均方根误差（RMSECV）为105.2 kg m−3。关于MFA，r2为0.77，RMSECV为5.36°。成功地绘制了木材密度以及MFA在一个高的空间分辨率。因此，年生长轮特征的检测和异质木材质量方面的评价已被促进。通过观察早材（EW）和晚材（LW）之间的密度差异，并通过Mäule颜色反应（Mäule color reaction）对映射结果进行了目视检查，表明在MFA验证方面，CW中的木质素含量较高，因为CW具有较高的MFA值。
Abstract Wood density and microfibril angle (MFA) are strongly correlated with wood stiffness, swelling/shrinkage, and its anisotropy. Understanding the spatial distribution of these data is critical for solid timber applications. In this study, near-infrared (NIR) hyperspectral imaging has been calibrated for evaluation of wood density and MFA in an effective manner. Briefly, five wood samples collected from both normal wood (NW) and compression wood (CW) moieties of two different Cryptomeria japonica trees were analyzed. Partial least squares (PLS) regression analysis was performed to determine the relationship between X-ray densitometry data obtained by SilviScan and NIR spectra, and cross-validation (leave-one-out) approach served for prediction performances. The validation coefficient of determination (r2) between the predicted densities by the NIR technique and the X-ray data was 0.83 with a root mean squared error of cross-validation (RMSECV) of 105.2 kg m−3. Regarding MFA, the r2 was 0.77 and RMSECV 5.36°. Wood density was successfully maped as well as the MFA at a high spatial resolution. As a result, the detection of annual growth ring features and evaluation of aspects of heterogeneous wood quality has been facilitated. The mapping results were visually checked by looking at the difference between earlywood (EW) and latewood (LW) for density and by means of the Mäule color reaction indicating high lignin contents in CW in terms of MFA validation as CWs have high MFA values.