文献类型：期刊文献

中文题名：基于近红外光谱与误差反向传播神经网络技术的三种人工林木材识别研究

英文题名：Recognition of Three Types of Plantation Wood Species with Near Infrared Spectra Coupled with Back-Propagation Network

作者：庞晓宇[1,2] 杨忠[1,2] 吕斌[2] 贾东宇[2]

第一作者：庞晓宇

通信作者：Yang, Z[1];Yang, Z[2]

机构：[1]中国林业科学研究院林业新技术研究所;[2]中国林业科学研究院木材工业研究所

年份：2016

卷号：36

期号：11

起止页码：3552-3556

中文期刊名：光谱学与光谱分析

外文期刊名：Spectroscopy and Spectral Analysis

收录：CSTPCD;;EI(收录号：20164703047501);Scopus(收录号：2-s2.0-84995751138);WOS:【SCI-EXPANDED(收录号:WOS:000388251100019)】；北大核心:【北大核心2014】；CSCD:【CSCD2015_2016】；PubMed;

基金：国家自然科学基金项目(30800889;31370711)资助

语种：中文

中文关键词：BP网络;近红外光谱;SIMCA;分类;尾叶桉、马尾松、南方无性系I-72杨

外文关键词：BP network; Near infrared spectroscopy; SIMCA; Classification; Eucalyptus urophylla; Pinus massoniana; Populus×euramericana（Dode）Guineir cv; ＂San Martino＂（1-72/58）;

分类号：O657.3;S781

摘要：利用近红外光谱结合误差反向传播神经网络（BP）对三种人工林木材（尾叶桉、马尾松、南方无性系I-72杨）进行识别,探讨隐含层神经元个数、光谱预处理方法、光谱范围对BP网络模型的影响,并与SIMCA法所建模型做比较。结果表明：（1）BP网络结合全波段（780~2 500nm）近红外光谱数据建模,识别正确率达到97.78%,并确定隐含层神经元数为13;（2）全波段光谱建模比短波段（780~1 100nm）和长波段（1 100~2 500nm）光谱建模识别效果好,其识别正确率分别为97.78%,95.56%和96.67%,用一阶导数和二阶导数对全波段光谱进行预处理后,BP网络模型识别正确率分别为93.33%和71.11%;用多元散射校正（MSC）对全波段光谱进行预处理后,BP网络模型识别正确率为98.89%,（3）在三种波段（780~2 500,780~1 100和1 100~2 500nm）光谱建模的情况下,BP网络建模识别正确率分别为95.56%,96.67%和97.78%,SIMCA模型识别正确率分别为76.67%,81.11%和82.22%,BP网络建模比SIMCA法建模对三种人工林木材的识别正确率高。
In this study,the near infrared spectroscopy coupled with Back-Propagation（BP）network was used for the recognition of three kinds of plantation wood（Eucalyptus urophylla,Pinus massoniana,Populus×euramericana（Dode）Guineir cv.＂San Martino＂（1-72/58））.The study considered the effects of hidden layer neurons number,spectral pretreatment method and spectral regions on BP model,which are compared with SIMCA model simultaneously.The results showed that,（1）the recognition rate was 97.78% achieved by BP network model with hidden layer neurons number 13 and the spectral region of 780~2 500 nm.（2）BP model with spectral region of 780~2 500 nm was more robust than the other two BP models with spectral regions of 780~1 100 and 1 100~2 500 nm,of which recognition rates were 97.78%,95.56% and 96.67%,respectively.After the full spectra was pretreated with the first derivative and the second derivative methods,the recognition rates of BP models fell down to 93.33% and 71.11%.However,the recognition rate of BP model rose to 98.89% with the full spectra being pretreated by the multiplicative scatter correction（MSC）.（3）Compared with SIMCA models that recognition rates of three spectral regions（780~2 500,780~1 100 nm,and 1 100~2 500nm）were 76.67%,81.11% and 82.22% respectively,BP network work models had higher recognition rates.