基于近红外光谱的木荷木纤维解剖结构PLSR模型构建

doi:10.13287/j.1001-9332.202410.003

应用生态学报 ›› 2024, Vol. 35 ›› Issue (10): 2794-2802.doi: 10.13287/j.1001-9332.202410.003

基于近红外光谱的木荷木纤维解剖结构PLSR模型构建

林诚富^1,2, 邵文³, 王家燚⁴, 张蕊^2*, 马丽珍³, 黄少华³, 范辉华⁵, 周志春²

¹东北林业大学林学院, 哈尔滨 150040;
²中国林业科学研究院亚热带林业研究所/全省林木育种重点实验室, 杭州 311400;
³福建省建瓯市林业技术推广中心, 福建建瓯 353100;
⁴河北农业大学林学院, 河北保定 071000;
⁵福建省林业科学研究院, 福州 350012

收稿日期:2024-03-19 接受日期:2024-08-12 出版日期:2024-10-18 发布日期:2025-04-18
通讯作者: * E-mail: zhangruicaf@caf.ac.cn
作者简介:林诚富, 男, 1996年生, 硕士研究生。主要从事林木遗传育种研究。E-mail: 877582304@qq.com
基金资助:
浙江省农业(林木)新品种选育重大科技专项重点课题(2021C02070-9)、福建省林木种苗科技攻关七期项目(ZMGG-0703)和江西省林业科技创新项目(201919)

PLSR model based on near-infrared spectroscopy for the detection of wood fiber anatomy of Schima superba.

LIN Chengfu^1,2, SHAO Wen³, WANG Jiayi⁴, ZHANG Rui^2*, MA Lizhen³, HUANG Shaohua³, FAN Huihua⁵, ZHOU Zhichun²

¹School of Forestry, Northeast Forestry University, Harbin 150040, China;
²Research Institute of Subtropical Forestry, Chinese Academy of Forestry/Zhejiang Key Laboratory of Forest Genetics and Bree-ding, Hangzhou 311400, China;
³Extending Center for Forestry Science and Technology of Jian’ou City, Fujian Province, Jian’ou 353100, Fujian, China;
⁴College of Forestry, Hebei Agricultural University, Baoding 071000, Hebei, China;
⁵Fujian Academy of Forestry Sciences, Fuzhou 350012, China

Received:2024-03-19 Accepted:2024-08-12 Online:2024-10-18 Published:2025-04-18

摘要/Abstract

摘要： 为快速获取木荷木纤维表型数据以评估木材质量,对18年生20个种源100个材料使用便携式近红外光谱仪采集光谱数据,同时测定木纤维基本密度和解剖结构等9个指标,通过SNV、OSC和MSC预处理光谱数据,并用CARS筛选波长,建立PLSR模型。结果表明: 林地与室内光谱数据存在显著差异,两者的光谱数据相对独立。SNV、OSC和MSC三种预处理方法对模型的预测效果差异显著,其中,OSC在林地和室内多项木纤维表型结构特征光谱预处理上表现优异,模型的预测精度林地R²=0.47～0.78(平均0.63),室内R²=0.54～0.82 (平均0.71)。而SNV和MSC方法仅对林地数据建立壁腔比模型的预测效果较好,其余模型效果不佳。通过CARS方法筛选波长后,林地和室内数据构建的模型预测精度得到有效提升(R²=0.58和0.72)。在CARS前后各执行一次OSC时,林地和室内数据构建的模型预测精度可分别提升至0.68和0.84。OSC预处理和CARS方法可以有效提高木纤维解剖结构构建模型的精度。木纤维长、双壁厚、腔径、木材基本密度、腔宽比和壁腔比可先通过OSC结合CARS进行处理,在经过一次OSC处理后建立PLSR模型,模型预测精度R²在0.80～0.95,可以预测评估木荷类木纤维物理性质指标。

关键词: 近红外光谱技术, 光谱分析, PLSR, 木荷, 木纤维

Abstract: To rapidly acquire fiber phenotypic data for wood quality assessment, we used a portable NIR spectro-meter to collect spectral data in 100 individuals of Schima superba at 18-year-old of 20 different provenances, and simultaneously collected wood cores. Wood basic density and the anatomical structure of wood fiber were measured. The standard normal variate (SNV), orthogonal signal correction (OSC), and multiplicative scatter correction (MSC) methods were used for spectral preprocessing, the competitive adaptive reweighted sampling (CARS) method were used for wavelength selection, and the partial least squares regression (PLSR) model were established. The results showed a significant difference for the absolute reflectance data between forest and indoor environments, and the spectral data of which were relatively independent. SNV, OSC and MSC showed significant differences for predictive performance of the model. OSC had the excellent preprocessing capability in multiple cha-racteristics of wood fiber ether in forest and indoor environments. The predictive accuracy of the models with R² was 0.47-0.78 in forest (average=0.63), and R² was 0.54-0.82 in indoor environment (average=0.71). However, the SNV and MSC methods could not establish the models, except the fiber wall-cavity ratio from forest data. After wavelength selection through the CARS method, the predictive accuracy of the models was significantly improved using both forest and indoor data (R²=0.58 and 0.72, respectively). When performed OSC before and after CARS, the predictive accuracy of the models was improved to 0.68 and 0.84 respectively using forest and indoor data. The OSC and CARS could significantly improve the accuracy of the models for wood fiber anatomical structures. First OSC, then CARS, and finally OSC methods could be used to establish the PLSR model for fiber length, fiber cell wall thickness, fiber lumen diameter, wood basic density, fiber cavity-width ratio, and fiber wall-cavity ratio, and the R² ranged from 0.80 to 0.95. These models had effective predictive ability and accuracy to assess the physical properties of wood fibers of S. superba.

Key words: near-infrared spectroscopy, spectral analysis, PLSR, Schima superb, wood fiber

林诚富, 邵文, 王家燚, 张蕊, 马丽珍, 黄少华, 范辉华, 周志春. 基于近红外光谱的木荷木纤维解剖结构PLSR模型构建[J]. 应用生态学报, 2024, 35(10): 2794-2802.

LIN Chengfu, SHAO Wen, WANG Jiayi, ZHANG Rui, MA Lizhen, HUANG Shaohua, FAN Huihua, ZHOU Zhichun. PLSR model based on near-infrared spectroscopy for the detection of wood fiber anatomy of Schima superba.[J]. Chinese Journal of Applied Ecology, 2024, 35(10): 2794-2802.

参考文献

[1] 李少锋. 林木木材形成机制及材性改良研究进展. 温带林业研究, 2019, 2(2): 40-47
[2] Wheeler EA, Baas P. Wood identification: A review. IAWA Journal, 1998, 19: 241-264
[3] Gao MY, Qi DW, Mu HB, et al. A transfer residual neural network based on ResNet-34 for detection of wood knot defects. Forests, 2021, 12: 212
[4] Schimleck LR, Evans R, Ilic J. Estimation of Eucalyptus delegatensis wood properties by near infrared spectroscopy. Canadian Journal of Forest Research, 2001, 31: 1671-1675
[5] Downes GM, Touza M, Harwood C, et al. NIR detection of non-recoverable collapse in sawn boards of Eucalyptus globulus. European Journal of Wood and Wood Pro-ducts, 2014, 72: 563-570
[6] Yu L, Liang YL, Zhang YZ, et al. Mechanical properties of wood materials using near-infrared spectroscopy based on correlation local embedding and partial least-squares. Journal of Forestry Research, 2020, 31: 1053-1060
[7] Schimleck LR, de Matos JLM, Da Silva Oliveira JT, et al. Non-destructive estimation of pernambuco (Caesalpinia echinata) clear wood properties using near infrared spectroscopy. Journal of Near Infrared Spectroscopy, 2011, 19: 411-419
[8] 丁丽, 相玉红, 黄安民, 等. BP神经网络与近红外光谱定量预测杉木中的综纤维素、木质素、微纤丝角. 光谱学与光谱分析, 2009, 29(7): 1784-1787
[9] 孙柏玲, 柴宇博, 黄安民, 等. 近红外光谱法在慈竹微纤丝角和纤维长度分析中的应用. 光谱学与光谱分析, 2011, 31(12): 3251-3255
[10] Cheng JH, Sun DW. Partial least squares regression (PLSR) applied to NIR and HSI spectral data modeling to predict chemical properties of fish muscle. Food Engineering Reviews, 2017, 9: 36-49
[11] Ozaki Y, Huck C, Tsuchikawa S, et al. Near-infrared spectroscopy: Theory, spectral analysis, instrumentation, and applications// Ozaki Y, Morita S, Morisawa Y, eds. Spectral Analysis in the NIR Spectroscopy. Singapore: Springer, 2021: 63-82
[12] 武斌, 沈嘉棋, 汪鑫, 等. 主成分分析排序和模糊线性判别分析的生菜近红外光谱分类. 光谱学与光谱分析, 2022, 42(10): 3079-3083
[13] 罗批, 郭继昌, 李锵, 等. 基于偏最小二乘回归建模的探讨. 天津大学学报, 2002, 35(6): 783-786
[14] 姚允龙, 王欣, 谭霄鹏, 等. 基于PLSR的典型沼泽湿地植物叶片性状与光谱模型构建: 以三江国家级自然保护区为例. 地理科学, 2022, 42(9): 1638-1645
[15] Li YJ, Dong X, Sun Y, et al. An efficient method to reduce grain angle influence on NIR spectra for predicting extractives content from heartwood stem cores of Toona sinensis. Plant Methods, 2020, 16: 77
[16] Isaksson T, Næs T. The effect of multiplicative scatter correction (MSC) and linearity improvement in NIR spectroscopy. Applied Spectroscopy, 1988, 42: 1273-1284
[17] 曹晓兰, 邓梦洁, 汪佩佩. 基于PLSR的苎麻叶片含水量估测模型建立及优化. 激光生物学报, 2018, 27(5): 467-473
[18] Bin J, Ai FF, Fan W, et al. An efficient variable selection method based on variable permutation and model population analysis for multivariate calibration of NIR spectra. Chemometrics and Intelligent Laboratory Systems, 2016, 158: 1-13
[19] 江泽慧, 黄安民, 王斌. 木材不同切面的近红外光谱信息与密度快速预测. 光谱学与光谱分析, 2006, 26(6): 1034-1037
[20] Tong L, Zhang WB. Using Fourier transform near-infrared spectroscopy to predict the mechanical properties of thermally modified southern pine wood. Applied Spectroscopy, 2016, 70: 1676-1684
[21] 汪睿, 石兰兰, 王玉荣. 利用近红外光谱技术快速预测楸树木材抗弯性质. 光谱学与光谱分析, 2023, 43(2): 557-562
[22] 赵荣军, 张黎, 霍小梅, 等. 基于近红外光谱技术预测径/弦切面粗皮桉木材微纤丝角. 光谱学与光谱分析, 2010, 30(9): 2355-2359
[23] 王云鹏, 张蕊, 周志春, 等. 10年生木荷生长和材性性状家系变异及选择. 南京林业大学学报: 自然科学版, 2020, 44(5): 85-92
[24] 张蕊, 王艺, 金国庆, 等. 模拟氮沉降对低磷胁迫下3个种源木荷幼苗生长及叶片氮磷含量的影响. 林业科学, 2015, 51(4): 36-43
[25] 王淼, 吴国亮, 张蕊, 等. 木荷18年生种源生长、材性的地理种源变异. 应用生态学报, 2023, 34(9): 2337-2344
[26] 王家燚, 陈焕伟, 张蕊, 等. 木荷全同胞家系生长与分枝性状的遗传变异及效应分析. 浙江农林大学学报, 2023, 40(4): 738-746
[27] 徐肇友, 王云鹏, 肖纪军, 等. 不同产地木荷优树无性系表型多样性. 东北林业大学学报, 2021, 49(2): 5-10
[28] 文国卫, 黄秋良, 吕增伟, 等. 气候变化情境下木荷潜在地理分布及生态适宜性分析. 生态学报, 2023, 43(16): 6617-6626
[29] Wang YP, Zhang R, Zhou ZC. Radial variation of wood anatomical properties determines the demarcation of juvenile-mature wood in Schima superba. Forests, 2021, 12: 512
[30] 楚秀丽, 王艺, 金国庆, 等. 不同生境、初植密度及林龄木荷人工林生长、材性变异及林分分化. 林业科学, 2014, 50(6): 152-159
[31] 周志春, 范辉华, 金国庆, 等. 木荷地理遗传变异和优良种源初选. 林业科学研究, 2006, 19(6): 718-724
[32] 王秀花, 马丽珍, 马雪红, 等. 木荷人工林生长和木材基本密度. 林业科学, 2011, 47(7): 138-144
[33] 张萍, 周志春, 金国庆, 等. 木荷种源遗传多样性和种源区初步划分. 林业科学, 2006, 42(2): 38-42
[34] Mensah BJK, Luboš B, Prince CA, et al. Comparison of field and laboratory wet soil spectra in the vis-nir range for soil organic carbon prediction in the absence of laboratory dry measurements. Remote Sensing, 2020, 12: 3082
[35] Upadhyay V, Kandpal KC, Jaiswal S, et al. Revisit and optimisation of spectral data collection techniques from vegetation using hand held non-imaging spectroscopic sensor for minimising errors. Vibrational Spectroscopy, 2020, 111: 103159
[36] Rossel RV, Mcbratney AB. Laboratory evaluation of a proximal sensing technique for simultaneous measurement of soil clay and water content. Geoderma, 1998, 85: 19-39
[37] Baumgardner MF, Silva LF, Biehl LL, et al. Reflectance properties of soils. Advances in Agronomy, 1986, 38: 1-44
[38] Mouazen AM, De Baerdemaeker J, Ramon H. Towards development of on-line soil moisture content sensor using a fibre-type NIR spectrophotometer. Soil and Tillage Research, 2005, 80: 171-183
[39] Schimleck LR, Stürzenbecher R, Mora C, et al. Comparison of Pinus taeda L. wood property calibrations based on NIR spectra from the radial-longitudinal and radial-transverse faces of wooden strips. Holzforschung, 2005, 59: 214-218
[40] Liu TT, Wang ZL, Zhang T, et al. Classification of su-gar content of kiwi fruit based on deep learning and near infrared spectrum. Research Square, 2023, DOI: 10.21203/rs.3.rs-3799363/v1
[41] Liang L, Wei LL, Fang GG, et al. Prediction of holocellulose and lignin content of pulp wood feedstock using near infrared spectroscopy and variable selection. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2020, 225: 117515

基于近红外光谱的木荷木纤维解剖结构PLSR模型构建

PLSR model based on near-infrared spectroscopy for the detection of wood fiber anatomy of Schima superba.

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王淼, 吴国亮, 张蕊, 王家燚, 王云鹏, 黄大庄, 周志春. 木荷18年生种源生长、材性的地理种源变异 [J]. 应用生态学报, 2023, 34(9): 2337-2344.
[2]	姚甲宝, 楚秀丽, 周志春, 童建设, 王晖, 余家中. 不同养分环境下邻株竞争对木荷和杉木生长、细根形态及分布的影响 [J]. 应用生态学报, 2017, 28(5): 1441-1447.
[3]	姚甲宝, 楚秀丽, 周志春, 童建设, 王晖, 余家中. 不同养分环境下木荷种源生长和根系发育对邻株竞争响应的差异 [J]. 应用生态学报, 2017, 28(4): 1087-1093.
[4]	张振振1,2,赵平1**,倪广艳1,朱丽薇1,赵秀华1,2,赵培强1,2,牛俊峰1. 华南丘陵植被恢复先锋树种木荷与马占相思的水分利用 [J]. 应用生态学报, 2014, 25(4): 931-939.
[5]	吕妍1,2,郑泽梅1,2**,美丽班·马木提1,2,陆江1,2,王希华1,2. 增施氮磷肥对木荷林凋落物生产量及其养分的影响 [J]. 应用生态学报, 2013, 24(11): 3027-3034.
[6]	廖春华,张显峰**,刘羽. 基于多端元光谱分解的干旱区植被覆盖度遥感反演 [J]. 应用生态学报, 2012, 23(12): 3243-3249.
[7]	罗媛媛;袁金凤;沈国春;赵谷风;于明坚. 常绿阔叶林片段中木荷凋落叶分解速率及中小型土壤节肢动物群落的结构动态 [J]. 应用生态学报, 2010, 21(2): 265-271.
[8]	杨智杰,陈光水,谢锦升,杨玉盛. 杉木、木荷纯林及其混交林凋落物量和碳归还量 [J]. 应用生态学报, 2010, 21(09): 2235-2240.
[9]	林磊;周志春. 水分和磷素对木荷不同种源苗木生长和磷效率的影响 [J]. 应用生态学报, 2009, 20(11): 2617-2623.
[10]	林磊¹;周志春¹;范辉华²;金国庆¹;陈奕良³;洪桂木⁴. 木荷稳定碳同位素分辨率的种源差异 [J]. 应用生态学报, 2009, 20(04): 741-746 .
[11]	马雪红;周志春;张一;金国庆. 异质养分环境下不同树种觅养行为与光照条件的关系 [J]. 应用生态学报, 2008, 19(05): 961-968 .
[12]	林巧佳, 刘景宏, 杨桂娣, 黄彪. 纸浆废液中有机物资源的利用 [J]. 应用生态学报, 2005, 16(4): 698-702.
[13]	林巧佳；刘景宏；杨桂娣；黄彪. 纸浆废液中有机物资源的利用 [J]. 应用生态学报, 2005, 16(04): 698-702 .
[14]	潘开文, 张远彬. 四川中亚热带扁刺栲-华木荷群系不同演替阶段土壤含水率的研究 [J]. 应用生态学报, 2003, (2): 165-169.
[15]	潘开文, 张远彬. 四川中亚热带扁刺栲-华木荷群系不同演替阶段土壤含水率的研究 [J]. 应用生态学报, 2003, (2): 165-169.