多信息融合的马尾松苗木静态指标检测方法
|
摘要
获取马尾松形态指标和主要生理指标有助于筛选优良的造林苗木。利用先进的立体视觉3D重构技术和化学计量学方法(近红外光谱检测技术)实现马尾松苗木的静态指标精准测量。首先搭建软硬件实验平台,通过图像采集系统获取图像序列,在计算机上通过算法重构出马尾松苗木的3D模型,提取马尾松苗木的形态学指标;然后利用化学计量学方法挖掘近红外光谱中的有用信息,并采用偏最小二乘回归方法建立线性模型,从而达到苗木主要生理指标的快速、无损测定。所选取的研究方法得到的马尾松苗木质量指标与人工实际测量得到的数据具有很强的相关性。基于多信息融合的马尾松苗木质量检测方法精度高,可满足林业调查要求,并为马尾松苗木活力相关研究提供了新的思路。
As one of the main forestation species,masson pine is widely distributed in southern China. Obtaining masson pine morphological index and physiological index is helpful for screening good forestation seedlings.Morphological indicators reflect the external characteristics of masson pine,and physiological indicators reflect the internal physiological status of masson pine. The traditional method of masson pine testing is complicated,time-consuming and laborious. It is of great significance for the construction of modern forestry ecological civilization to quickly and accurately evaluate the quality of masson pine seedlings by means of advanced technologies. A newmethod based on advanced stereo vision 3 D reconstruction technology and the stoichiometric method( near infrared spectroscopy) detection technology to evaluate the quality of masson pine was proposed in this paper. Firstly,the software and hardware platform were built to reconstruct the three-dimensional model of masson pine seedlings by multi viewimage sequence,and the morphological parameters of the seedlings were extracted quickly. Then the useful information in the near infrared spectrum was excavated by the chemometrics method and the spectral model was established to achieve the rapid and nondestructive measurement of the physiological indexes of the seedlings. The algorithm could optimize the characteristic wave number of near infrared spectrum and improve the accuracy of measurement. The software extracted values and manually measured values of crown,trunk and root length and volume of masson pine seedlings were linear fitted. The quality index of masson pine seedlings selected by the research method was highly correlated with the data obtained by manual measurement. The results showed that thequality detection method of masson pine seedlings based on multi information fusion could meet the requirements of forestry investigation,save time and improve the accuracy,and provide a newway of thinking for the study of the vitality of masson pine. Based on this study,further research directions in the future were proposed.
As one of the main forestation species,masson pine is widely distributed in southern China. Obtaining masson pine morphological index and physiological index is helpful for screening good forestation seedlings.Morphological indicators reflect the external characteristics of masson pine,and physiological indicators reflect the internal physiological status of masson pine. The traditional method of masson pine testing is complicated,time-consuming and laborious. It is of great significance for the construction of modern forestry ecological civilization to quickly and accurately evaluate the quality of masson pine seedlings by means of advanced technologies. A newmethod based on advanced stereo vision 3 D reconstruction technology and the stoichiometric method( near infrared spectroscopy) detection technology to evaluate the quality of masson pine was proposed in this paper. Firstly,the software and hardware platform were built to reconstruct the three-dimensional model of masson pine seedlings by multi viewimage sequence,and the morphological parameters of the seedlings were extracted quickly. Then the useful information in the near infrared spectrum was excavated by the chemometrics method and the spectral model was established to achieve the rapid and nondestructive measurement of the physiological indexes of the seedlings. The algorithm could optimize the characteristic wave number of near infrared spectrum and improve the accuracy of measurement. The software extracted values and manually measured values of crown,trunk and root length and volume of masson pine seedlings were linear fitted. The quality index of masson pine seedlings selected by the research method was highly correlated with the data obtained by manual measurement. The results showed that thequality detection method of masson pine seedlings based on multi information fusion could meet the requirements of forestry investigation,save time and improve the accuracy,and provide a newway of thinking for the study of the vitality of masson pine. Based on this study,further research directions in the future were proposed.
引文
[1]刘君昂,周国英,靳爱仙,等.马尾松人工林健康评价及生态调控关键技术研究[M].杨凌:西北农林科技大学出版社,2013.LIU J A,ZHOU G Y,JIN A X,et al. Key technology of health evaluation and ecological regulation on Pinus massonian plantation[M]. Yangling:Northwest A&F University Press Co.,Ltd.,2013.
[2]李国雷,刘勇,祝燕,等.国外苗木质量研究进展[J].世界林业研究,2011,24(2):27-35. DOI:10.13348/j.cnki.sjlyyj.2011.02.009.LI G L,LIU Y,ZHU Y,et al. A review of oversea studies of seedling quality[J]. World Forestry Research,2011,24(2):27-35.
[3]李国雷,刘勇,祝燕,等.国外容器苗质量调控技术研究进展[J].林业科学,2012,48(8):135-142. DOI:10. 11707/j.1001-7488.20120822.LI G L,LIU Y,ZHU Y,et al. A review on the abroad studies of techniques in regulating quality of container seedling[J]. Scientia Silvae Sinicae,2012,48(8):135-142.
[4]赵春江,陆声链,郭新宇,等.数字植物研究进展:植物形态结构三维数字化[J].中国农业科学,2015,48(17):3415-3428. DOI:10.3864/j.issn.0578-1752.2015.17.009.ZHAO C J,LU S L,GUO X Y,et al. Advances in research of digital plant:3D digitization of plant morphological structure[J].Scientia Agricultura Sinica,2015,48(17):3415-3428.
[5]袁晓敏,赵春江,温维亮,等.番茄植株三维形态精确重构研究[J].农业机械学报,2012,43(12):204-210. DOI:10.6041/j.issn.1000-1298.2012.12.037.YUAN X M,ZHAO C J,WEN W L,et al. Detailed modeling of3-d configuration of tomato plant[J]. Transactions of the Chinese Society for Agricultural Machinery,2012,43(12):204-210.
[6]张建,李宗南,张楠,等.基于实测数据的作物三维信息获取与重建方法研究进展[J].华中农业大学学报,2013,32(4):126-134. DOI:10.3969/j.issn.1000-2421.2013.04.023.ZHANG J, LI Z N, ZHANG N, et al. Advances in 3D information collection and reconstruction of crop based on the measured data[J]. Journal of Huazhong Agricultural University,2013,32(4):126-134.
[7]温维亮,郭新宇,王勇健,等.葡萄树地上部形态结构数据获取方法[J].农业工程学报,2015,31(22):161-168. DOI:10.11975/j.issn.1002-6819.2015.22.022.WEN W L,GUO X N,WANG Y J,et al. Morphological and structural data acquisition for above-ground part of grapevine[J].Transactions of the Chinese Society of Agricultural Engineering,2015,31(22):161-168.
[8]倪超,张云,高捍东,等.基于多目视觉的马尾松苗木形态学参数提取研究[J].林业工程学报,2018,3(2):123-128.DOI:10.13360/j.issn.2096-1359.2018.02.020.NI C,ZHANG Y,GAO H D,et al. Study on extraction of morphological parameters of masson pine seedlings based on multiview stereo vision[J]. Journal of Forestry Engineering,2018,3(2):123-128.
[9]ADHIKARI B,KARKEE M. 3D reconstruction of apple trees for mechanical pruning[C]//2011 ASABE Annual Meeting,ASABE Paper 1111613,2011.
[10]WU C Y,HE L Y,DU X Q,et al. 3D reconstruction of Chinese hickory tree for dynamics analysis[J]. Biosystems Engineering,2014,119:69-79. DOI:10.1016/j.biosystemseng.2014.01.008.
[11]ULLAH S,SKIDMORE A K,GROEN T A,et al. Evaluation of three proposed indices for the retrieval of leaf water content from the mid-wave infrared(2-6μm)spectra[J]. Agricultural and Forest Meteorology,2013,171/172:65-71. DOI:10. 1016/j.agrformet.2012.11.014.
[12]CHENG T,RIVARD B,SNCHEZ-AZOFEIFA A G,et al. Predicting leaf gravimetric water content from foliar reflectance across a range of plant species using continuous wavelet analysis[J].Journal of Plant Physiology,2012,169(12):1134-1142. DOI:10.1016/j.jplph.2012.04.006.
[13]LI Y X,ZHANG Y Z,JIANG L C. Modeling chlorophyll content of Korean Pine needles with NIR and SVM[J]. Procedia Environmental Sciences,2011, 10:222-227. DOI:10. 1016/j.proenv.2011.09.038.
[14]ECARNOT M,COMPAN F,ROUMET P. Assessing leaf nitrogen content and leaf mass per unit area of wheat in the field throughout plant cycle with a portable spectrometer[J]. Field Crops Research,2013,140:44-50. DOI:10. 1016/j. fcr. 2012.10.013.
[15]XU L,SHI P T,YE Z H,et al. Rapid analysis of adulterations in Chinese lotus root powder(LRP)by near-infrared(NIR)spectroscopy coupled with chemometric class modeling techniques[J]. Food Chemistry,2013,141(3):2434-2439. DOI:10.1016/j.foodchem.2013.05.104.
[16]李跑,吴红艳,李尚科,等.近红外光谱技术结合化学计量方法用于大米的快速分析[J].食品研究与开发,2018,39(19):117-124. DOI:3969/j.issn.1005-6521.2018.19.021.LI P,WU H Y,LI S K,et al. Near infrared spectroscopy combined with chemometrics method for rapid analysis of rice[J].Food Research and Development,2018,39(19):117-124.
[17]HAASE D L. Understanding forest seedling quality:measurements and interpretation[J]. Tree Planters Notes,2008,52(2):24-30.
[2]李国雷,刘勇,祝燕,等.国外苗木质量研究进展[J].世界林业研究,2011,24(2):27-35. DOI:10.13348/j.cnki.sjlyyj.2011.02.009.LI G L,LIU Y,ZHU Y,et al. A review of oversea studies of seedling quality[J]. World Forestry Research,2011,24(2):27-35.
[3]李国雷,刘勇,祝燕,等.国外容器苗质量调控技术研究进展[J].林业科学,2012,48(8):135-142. DOI:10. 11707/j.1001-7488.20120822.LI G L,LIU Y,ZHU Y,et al. A review on the abroad studies of techniques in regulating quality of container seedling[J]. Scientia Silvae Sinicae,2012,48(8):135-142.
[4]赵春江,陆声链,郭新宇,等.数字植物研究进展:植物形态结构三维数字化[J].中国农业科学,2015,48(17):3415-3428. DOI:10.3864/j.issn.0578-1752.2015.17.009.ZHAO C J,LU S L,GUO X Y,et al. Advances in research of digital plant:3D digitization of plant morphological structure[J].Scientia Agricultura Sinica,2015,48(17):3415-3428.
[5]袁晓敏,赵春江,温维亮,等.番茄植株三维形态精确重构研究[J].农业机械学报,2012,43(12):204-210. DOI:10.6041/j.issn.1000-1298.2012.12.037.YUAN X M,ZHAO C J,WEN W L,et al. Detailed modeling of3-d configuration of tomato plant[J]. Transactions of the Chinese Society for Agricultural Machinery,2012,43(12):204-210.
[6]张建,李宗南,张楠,等.基于实测数据的作物三维信息获取与重建方法研究进展[J].华中农业大学学报,2013,32(4):126-134. DOI:10.3969/j.issn.1000-2421.2013.04.023.ZHANG J, LI Z N, ZHANG N, et al. Advances in 3D information collection and reconstruction of crop based on the measured data[J]. Journal of Huazhong Agricultural University,2013,32(4):126-134.
[7]温维亮,郭新宇,王勇健,等.葡萄树地上部形态结构数据获取方法[J].农业工程学报,2015,31(22):161-168. DOI:10.11975/j.issn.1002-6819.2015.22.022.WEN W L,GUO X N,WANG Y J,et al. Morphological and structural data acquisition for above-ground part of grapevine[J].Transactions of the Chinese Society of Agricultural Engineering,2015,31(22):161-168.
[8]倪超,张云,高捍东,等.基于多目视觉的马尾松苗木形态学参数提取研究[J].林业工程学报,2018,3(2):123-128.DOI:10.13360/j.issn.2096-1359.2018.02.020.NI C,ZHANG Y,GAO H D,et al. Study on extraction of morphological parameters of masson pine seedlings based on multiview stereo vision[J]. Journal of Forestry Engineering,2018,3(2):123-128.
[9]ADHIKARI B,KARKEE M. 3D reconstruction of apple trees for mechanical pruning[C]//2011 ASABE Annual Meeting,ASABE Paper 1111613,2011.
[10]WU C Y,HE L Y,DU X Q,et al. 3D reconstruction of Chinese hickory tree for dynamics analysis[J]. Biosystems Engineering,2014,119:69-79. DOI:10.1016/j.biosystemseng.2014.01.008.
[11]ULLAH S,SKIDMORE A K,GROEN T A,et al. Evaluation of three proposed indices for the retrieval of leaf water content from the mid-wave infrared(2-6μm)spectra[J]. Agricultural and Forest Meteorology,2013,171/172:65-71. DOI:10. 1016/j.agrformet.2012.11.014.
[12]CHENG T,RIVARD B,SNCHEZ-AZOFEIFA A G,et al. Predicting leaf gravimetric water content from foliar reflectance across a range of plant species using continuous wavelet analysis[J].Journal of Plant Physiology,2012,169(12):1134-1142. DOI:10.1016/j.jplph.2012.04.006.
[13]LI Y X,ZHANG Y Z,JIANG L C. Modeling chlorophyll content of Korean Pine needles with NIR and SVM[J]. Procedia Environmental Sciences,2011, 10:222-227. DOI:10. 1016/j.proenv.2011.09.038.
[14]ECARNOT M,COMPAN F,ROUMET P. Assessing leaf nitrogen content and leaf mass per unit area of wheat in the field throughout plant cycle with a portable spectrometer[J]. Field Crops Research,2013,140:44-50. DOI:10. 1016/j. fcr. 2012.10.013.
[15]XU L,SHI P T,YE Z H,et al. Rapid analysis of adulterations in Chinese lotus root powder(LRP)by near-infrared(NIR)spectroscopy coupled with chemometric class modeling techniques[J]. Food Chemistry,2013,141(3):2434-2439. DOI:10.1016/j.foodchem.2013.05.104.
[16]李跑,吴红艳,李尚科,等.近红外光谱技术结合化学计量方法用于大米的快速分析[J].食品研究与开发,2018,39(19):117-124. DOI:3969/j.issn.1005-6521.2018.19.021.LI P,WU H Y,LI S K,et al. Near infrared spectroscopy combined with chemometrics method for rapid analysis of rice[J].Food Research and Development,2018,39(19):117-124.
[17]HAASE D L. Understanding forest seedling quality:measurements and interpretation[J]. Tree Planters Notes,2008,52(2):24-30.