基于随机森林回归算法的苹果树冠层光照分布模型
|
摘要
合理的果树冠层结构有利于光照的有效分布,对提升果实产量与品质有重要意义。为揭示果树冠层内部的光照分布情况,针对目前果树冠层内部光照强度获取难度大、预测精度低的问题,研究了冠层颜色特征与光照强度的对应关系,提出一种基于冠层剖面阴影特征和冠层点云颜色特征的随机森林预测模型。以纺锤形"陕富6号"苹果树为研究对象,首先使用Kinect 2. 0采集果树的双面点云数据,预处理后得到完整的点云数据;其次,基于改进的空间殖民算法和叶序添加规则重构果树的三维模型;最后,使用"切片法",在垂直方向上将冠层模型每0. 1 m分层划分,使用POV-Ray渲染器逐层渲染阴影,同时使用光照度计,自顶向下每0. 1 m实测光照强度数据,构建以每层阴影图灰度特征和每层点云HSI颜色特征为输入,以相对光照强度为输出的随机森林网络。试验结果表明,该方法能够较为准确地预测冠层内的光照分布情况,预测值与实际值的决定系数R~2为0. 864,平均绝对百分比误差MAPE为0. 236,RF回归模型可作为苹果树冠层内光照分布预测的有效方法,为果树的剪枝、整形等研究提供参考。
The reasonable structure of fruit tree canopy is beneficial to the effective distribution of illumination,which has vital significance to enhance the fruit yield and quality. At present,it is difficult to obtain illumination intensity data in the canopy of fruit trees,and the prediction accuracy is low. In order to study inner canopy illumination distribution,a random forest prediction model was proposed based on canopy profile shadow feature and point cloud color feature. The detailed research methods were shown as follows. Firstly,the spindle "Shanfu 6 "apple tree was chosen as the research object and Kinect 2. 0 was used to acquire double face point cloud data of tree,and then the complete data was obtained with pre-process. Secondly, the improved space colonization algorithm with growth angle constraint and phyllotaxis adding rules were used to rebuild apple tree 3D model. Finally,the "slice method"was used to cut canopy model every 0. 1 m in the vertical direction,and then the POV-Ray renderer was used to render shadows layer after layer,meanwhile,light meter was used to obtain illumination intensity data every 0. 1 m from top to bottom consistently,and the random forest network that with input data of color feature of every layer and output data of relative illumination intensity was built as the apple tree canopy illumination distribution prediction model. The experiment results showed that the proposed method can predict the illumination distribution accurately. The determination coefficient R~2 between true value and predicted value was 0. 864,and MAPE was 0. 236. Random forest regression model can be used as an efficient method for prediction of canopy illumination distribution,and it can provide reference for fruit tree pruning and plastic research.
The reasonable structure of fruit tree canopy is beneficial to the effective distribution of illumination,which has vital significance to enhance the fruit yield and quality. At present,it is difficult to obtain illumination intensity data in the canopy of fruit trees,and the prediction accuracy is low. In order to study inner canopy illumination distribution,a random forest prediction model was proposed based on canopy profile shadow feature and point cloud color feature. The detailed research methods were shown as follows. Firstly,the spindle "Shanfu 6 "apple tree was chosen as the research object and Kinect 2. 0 was used to acquire double face point cloud data of tree,and then the complete data was obtained with pre-process. Secondly, the improved space colonization algorithm with growth angle constraint and phyllotaxis adding rules were used to rebuild apple tree 3D model. Finally,the "slice method"was used to cut canopy model every 0. 1 m in the vertical direction,and then the POV-Ray renderer was used to render shadows layer after layer,meanwhile,light meter was used to obtain illumination intensity data every 0. 1 m from top to bottom consistently,and the random forest network that with input data of color feature of every layer and output data of relative illumination intensity was built as the apple tree canopy illumination distribution prediction model. The experiment results showed that the proposed method can predict the illumination distribution accurately. The determination coefficient R~2 between true value and predicted value was 0. 864,and MAPE was 0. 236. Random forest regression model can be used as an efficient method for prediction of canopy illumination distribution,and it can provide reference for fruit tree pruning and plastic research.
引文
[1]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.ZHAO Chunjiang,LU Shenglian,GUO Xinyu,et al.Advances in research of digital plant:3D digitization of plant morphological structure[J].Scientia Agricultura Sinica,2015,48(17):3415-3428.(in Chinese)
[2]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.ZHANG Jian,LI Zongnan,ZHANG Nan,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.(in Chinese)
[3]HONDA H.Description of the form of trees by the parameters of the tree-like body:effects of the branching angle and the branch length on the shape of the tree-like body[J].Journal of Theoretical Biology,1971(31):331-338.
[4]章兰芬,李丙智,张社奇,等.基于三维数字化仪的高纺锤形苹果树光截获率测定[J].北方园艺,2012(7):9-12.ZHANG Lanfen,LI Bingzhi,ZHANG Sheqi,et al.Light interception measurement of apple trees trained to tall spindle shape base on three-dimensional digitizer[J].Northern Horticulture,2012(7):9-12.(in Chinese)
[5]PRUSINKIEWICZ P,LINDENMAYAR A.The algorithmic beauty of plants[M].New York:Springer-Verlag Press,1990.
[6]张晶,师翊,胡少军,等.基于Ray Tracing的数字植株冠层光照分布研究[J].农机化研究,2019,41(1):26-33.ZHANG Jing,SHI Yi,HU Shaojun,et al.Research of illumination distribution of digital plant canopy based on Ray Tracing[J].Journal of Agricultural Mechanization Research,2019,41(1):26-33.(in Chinese)
[7]郭彩玲,张伟洁,刘刚,等.基于苹果树冠层计盒维数的光照分布预测[J].农业工程学报,2018,34(16):177-183.GUO Cailing,ZHANG Weijie,LIU Gang,et al.Illumination spatial distribution prediction method based on apple tree canopy box-counting dimension[J].Transactions of the CSAE,2018,34(16):177-183.(in Chinese)
[8]JAROSZ W.Efficient Monte Carlo methods for light transport in scattering media[M].San Diego:University of California at San Diego,2008.
[9]马晓丹,郭彩玲,张雪,等.基于三维点云颜色特征的苹果树冠层光照分布计算方法[J/OL].农业机械学报,2015,46(6):263-268.MA Xiaodan,GUO Cailing,ZHANG Xue,et al.Calculation of light distribution of apple tree canopy based on color characteristics of 3D point cloud[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(6):263-268.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20150638&flag=1.DOI:10.6041/j.issn.1000-1298.2015.06.038.(in Chinese)
[10]WANG N,SHEN Y,HUA J,et al.Analyzing the canopy light distribution among different poplar genotypes using terrestrial laser scanner and the Green Lab model[C]∥International Conference on Functional Structural Plant Growth Modeling,Simulation,Visualization and Applications.IEEE,2017:214-223.
[11]GOVIND A.On the nature of canopy illumination due to differences in elemental orientation and aggregation for radiative transfer[J].International Journal of Biometeorology,2014,58(8):1803-1809.
[12]NASIR A K,TAJ M,KHAN M F.Evaluation of microsoft kinect sensor for plant health monitoring[J].IFAC-Papers on Line,2016,49(16):221-225.
[13]YIN B,SHAH D,TANG L.3D perception-based collision-free robotic leaf probing for automated indoor plant phenotyping[J].Transactions of the ASABE,2018,61(3):859-872.
[14]姚吉利.3维坐标转换参数直接计算的严密公式[J].测绘通报,2006(5):7-10.YAO Jili.Rigorous formula for direct calculating parameter in 3D transformation[J].Bulletin of Surveying and Mapping,2006(5):7-10.(in Chinese)
[15]BICHO A D L,RAFAEL A R,MUSSE S R,et al.Simulating crowds based on a space colonization algorithm[J].Computers&Graphics,2012,36(2):70-79.
[16]SHINOZAKI K,YODA K,HOZUMI K,et al.A quantitative analysis of plant form-the pipe model theoryⅠ.Basic analyses[J].Japanese Journal of Ecology,1964,14(3):97-105.
[17]SHINOZAKI K,YODA K,HOZUMI K,et al.A quantitative analysis of plant form-the pipe model theoryⅡ.Further evidence of the theory and its application in forest ecology[J].Japanese Journal of Ecology,1964,14(4):113-139.
[18]XIE K,YAN F,SHARF A,et al.Tree modeling with real tree-parts examples[J].IEEE Transactions on Visualization and Computer Graphics,2016,22(12):2608-2618.
[19]胡少军,耿楠,张志毅,等.基于稀疏图像的真实树交互建模方法[J].农业工程学报,2014,30(9):168-175.HU Shaojun,GENG Nan,ZHANG Zhiyi,et al.Interactive modeling of outdoor trees based on spare images[J].Transactions of the CSAE,2014,30(9):168-175.(in Chinese)
[20]LETCHFORD A N.A new separation algorithm for the Boolean quadric and cut polytopes[J].Discrete Optimization,2014,14:61-71.
[21]O'KEEFFE S E,BRELAND J,SHIRATUDDIN M F.MPI and POV-Ray parallel solution to improve performance of photorealistic rendering utilizing albacore and sequoia[C]∥International Conference on Computer&Information Science,2012.
[22]张徐洲,杜朋朋,何勇,等.植物BRDF研究及应用进展[J].光谱学与光谱分析,2017,37(3):829-835.ZHANG Xuzhou,DU Pengpeng,HE Yong,et al.Review of research and application for vegetation BRDF[J].Spectroscopy and Spectral Analysis,2017,37(3):829-835.(in Chinese)
[23]杨丽丽,张大卫,谢锐,等.初果期苹果树剪枝仿真研究[J/OL].农业机械学报,2017,48(增刊):98-102,333.YANG Lili,ZHANG Dawei,XIE Rui,et al.Study on pruning simulation of apple trees at initial fruit stage[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2017,48(Supp.):98-102,333.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=2017s016&flag=1.DOI:10.6041/j.issn.1000-1298.2017.S0.016.(in Chinese)
[24]刘伟峰,谢永杰,陈若望,等.天顶亮度与太阳高度角关系的观测[J].光电工程,2012,39(7):49-54.LIU Weifeng,XIE Yongjie,CHEN Ruowang,et al.Observation of relationship between zenith luminance and sun high angle[J].Opto-Electronic Engineering,2012,39(7):49-54.(in Chinese)
[25]马晓丹.苹果树冠层光照分布计算方法研究[D].北京:中国农业大学,2015.MA Xiaodan.Calculation of light distribution to apple tree canopy[D].Beijing:China Agricultural University,2015.(in Chinese)
[26]张蓉鑫,邢艳秋,张新伟,等.ICESat-GLAS激光天顶角对反演森林冠层高度的影响[J].中南林业科技大学学报,2018,38(9):26-33.ZHANG Rongxin,XING Yanqiu,ZHANG Xinwei,et al.Effect of laser off-nadir angle on estimating forest canopy height[J].Journal of Central South University of Forestry&Technology,2018,38(9):26-33.(in Chinese)
[27]张晶.苹果树冠层光照模型关键技术研究[D].杨凌:西北农林科技大学,2018.ZHANG Jing.Research on key techniques of illumination model on the canopy of apple trees[D].Yangling:Northwest A&FUniversity,2018.(in Chinese)
[28]牛亚晓,张立元,韩文霆.基于Lab颜色空间的棉花覆盖度提取方法研究[J/OL].农业机械学报,2018,49(10):240-249.NIU Yaxiao,ZHANG Liyuan,HAN Wenting.Extraction methods of cotton coverage based on Lab color space[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2018,49(10):240-249.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20181027&flag=1.DOI:10.6041/j.issn.1000-1298.2018.10.027.(in Chinese)
[29]王丽爱,马昌,周旭东,等.基于随机森林回归算法的小麦叶片SPAD值遥感估算[J/OL].农业机械学报,2015,46(1):259-265.WANG Liai,MA Chang,ZHOU Xudong,et al.Estimation of wheat leaf SPAD value using RF algorithmic model and remote sensing data[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(1):259-265.http:∥www.jcsam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20150136&flag=1.DOI:10.6041/j.issn.1000-1298.2015.01.036.(in Chinese)
[30]陈锡龙.黄土高原地区高纺锤苹果树冠层3D模拟光截获与光合产物积累研究[D].杨凌:西北农林科技大学,2013.CHEN Xilong.An experimental study of light interception and photosynthesis product for 3D virtual tall-spindle apple canopy in Loess Plateau region of China[D].Yangling:Northwest A&F University,2013.(in Chinese)
[2]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.ZHANG Jian,LI Zongnan,ZHANG Nan,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.(in Chinese)
[3]HONDA H.Description of the form of trees by the parameters of the tree-like body:effects of the branching angle and the branch length on the shape of the tree-like body[J].Journal of Theoretical Biology,1971(31):331-338.
[4]章兰芬,李丙智,张社奇,等.基于三维数字化仪的高纺锤形苹果树光截获率测定[J].北方园艺,2012(7):9-12.ZHANG Lanfen,LI Bingzhi,ZHANG Sheqi,et al.Light interception measurement of apple trees trained to tall spindle shape base on three-dimensional digitizer[J].Northern Horticulture,2012(7):9-12.(in Chinese)
[5]PRUSINKIEWICZ P,LINDENMAYAR A.The algorithmic beauty of plants[M].New York:Springer-Verlag Press,1990.
[6]张晶,师翊,胡少军,等.基于Ray Tracing的数字植株冠层光照分布研究[J].农机化研究,2019,41(1):26-33.ZHANG Jing,SHI Yi,HU Shaojun,et al.Research of illumination distribution of digital plant canopy based on Ray Tracing[J].Journal of Agricultural Mechanization Research,2019,41(1):26-33.(in Chinese)
[7]郭彩玲,张伟洁,刘刚,等.基于苹果树冠层计盒维数的光照分布预测[J].农业工程学报,2018,34(16):177-183.GUO Cailing,ZHANG Weijie,LIU Gang,et al.Illumination spatial distribution prediction method based on apple tree canopy box-counting dimension[J].Transactions of the CSAE,2018,34(16):177-183.(in Chinese)
[8]JAROSZ W.Efficient Monte Carlo methods for light transport in scattering media[M].San Diego:University of California at San Diego,2008.
[9]马晓丹,郭彩玲,张雪,等.基于三维点云颜色特征的苹果树冠层光照分布计算方法[J/OL].农业机械学报,2015,46(6):263-268.MA Xiaodan,GUO Cailing,ZHANG Xue,et al.Calculation of light distribution of apple tree canopy based on color characteristics of 3D point cloud[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(6):263-268.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20150638&flag=1.DOI:10.6041/j.issn.1000-1298.2015.06.038.(in Chinese)
[10]WANG N,SHEN Y,HUA J,et al.Analyzing the canopy light distribution among different poplar genotypes using terrestrial laser scanner and the Green Lab model[C]∥International Conference on Functional Structural Plant Growth Modeling,Simulation,Visualization and Applications.IEEE,2017:214-223.
[11]GOVIND A.On the nature of canopy illumination due to differences in elemental orientation and aggregation for radiative transfer[J].International Journal of Biometeorology,2014,58(8):1803-1809.
[12]NASIR A K,TAJ M,KHAN M F.Evaluation of microsoft kinect sensor for plant health monitoring[J].IFAC-Papers on Line,2016,49(16):221-225.
[13]YIN B,SHAH D,TANG L.3D perception-based collision-free robotic leaf probing for automated indoor plant phenotyping[J].Transactions of the ASABE,2018,61(3):859-872.
[14]姚吉利.3维坐标转换参数直接计算的严密公式[J].测绘通报,2006(5):7-10.YAO Jili.Rigorous formula for direct calculating parameter in 3D transformation[J].Bulletin of Surveying and Mapping,2006(5):7-10.(in Chinese)
[15]BICHO A D L,RAFAEL A R,MUSSE S R,et al.Simulating crowds based on a space colonization algorithm[J].Computers&Graphics,2012,36(2):70-79.
[16]SHINOZAKI K,YODA K,HOZUMI K,et al.A quantitative analysis of plant form-the pipe model theoryⅠ.Basic analyses[J].Japanese Journal of Ecology,1964,14(3):97-105.
[17]SHINOZAKI K,YODA K,HOZUMI K,et al.A quantitative analysis of plant form-the pipe model theoryⅡ.Further evidence of the theory and its application in forest ecology[J].Japanese Journal of Ecology,1964,14(4):113-139.
[18]XIE K,YAN F,SHARF A,et al.Tree modeling with real tree-parts examples[J].IEEE Transactions on Visualization and Computer Graphics,2016,22(12):2608-2618.
[19]胡少军,耿楠,张志毅,等.基于稀疏图像的真实树交互建模方法[J].农业工程学报,2014,30(9):168-175.HU Shaojun,GENG Nan,ZHANG Zhiyi,et al.Interactive modeling of outdoor trees based on spare images[J].Transactions of the CSAE,2014,30(9):168-175.(in Chinese)
[20]LETCHFORD A N.A new separation algorithm for the Boolean quadric and cut polytopes[J].Discrete Optimization,2014,14:61-71.
[21]O'KEEFFE S E,BRELAND J,SHIRATUDDIN M F.MPI and POV-Ray parallel solution to improve performance of photorealistic rendering utilizing albacore and sequoia[C]∥International Conference on Computer&Information Science,2012.
[22]张徐洲,杜朋朋,何勇,等.植物BRDF研究及应用进展[J].光谱学与光谱分析,2017,37(3):829-835.ZHANG Xuzhou,DU Pengpeng,HE Yong,et al.Review of research and application for vegetation BRDF[J].Spectroscopy and Spectral Analysis,2017,37(3):829-835.(in Chinese)
[23]杨丽丽,张大卫,谢锐,等.初果期苹果树剪枝仿真研究[J/OL].农业机械学报,2017,48(增刊):98-102,333.YANG Lili,ZHANG Dawei,XIE Rui,et al.Study on pruning simulation of apple trees at initial fruit stage[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2017,48(Supp.):98-102,333.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=2017s016&flag=1.DOI:10.6041/j.issn.1000-1298.2017.S0.016.(in Chinese)
[24]刘伟峰,谢永杰,陈若望,等.天顶亮度与太阳高度角关系的观测[J].光电工程,2012,39(7):49-54.LIU Weifeng,XIE Yongjie,CHEN Ruowang,et al.Observation of relationship between zenith luminance and sun high angle[J].Opto-Electronic Engineering,2012,39(7):49-54.(in Chinese)
[25]马晓丹.苹果树冠层光照分布计算方法研究[D].北京:中国农业大学,2015.MA Xiaodan.Calculation of light distribution to apple tree canopy[D].Beijing:China Agricultural University,2015.(in Chinese)
[26]张蓉鑫,邢艳秋,张新伟,等.ICESat-GLAS激光天顶角对反演森林冠层高度的影响[J].中南林业科技大学学报,2018,38(9):26-33.ZHANG Rongxin,XING Yanqiu,ZHANG Xinwei,et al.Effect of laser off-nadir angle on estimating forest canopy height[J].Journal of Central South University of Forestry&Technology,2018,38(9):26-33.(in Chinese)
[27]张晶.苹果树冠层光照模型关键技术研究[D].杨凌:西北农林科技大学,2018.ZHANG Jing.Research on key techniques of illumination model on the canopy of apple trees[D].Yangling:Northwest A&FUniversity,2018.(in Chinese)
[28]牛亚晓,张立元,韩文霆.基于Lab颜色空间的棉花覆盖度提取方法研究[J/OL].农业机械学报,2018,49(10):240-249.NIU Yaxiao,ZHANG Liyuan,HAN Wenting.Extraction methods of cotton coverage based on Lab color space[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2018,49(10):240-249.http:∥www.j-csam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20181027&flag=1.DOI:10.6041/j.issn.1000-1298.2018.10.027.(in Chinese)
[29]王丽爱,马昌,周旭东,等.基于随机森林回归算法的小麦叶片SPAD值遥感估算[J/OL].农业机械学报,2015,46(1):259-265.WANG Liai,MA Chang,ZHOU Xudong,et al.Estimation of wheat leaf SPAD value using RF algorithmic model and remote sensing data[J/OL].Transactions of the Chinese Society for Agricultural Machinery,2015,46(1):259-265.http:∥www.jcsam.org/jcsam/ch/reader/view_abstract.aspx?file_no=20150136&flag=1.DOI:10.6041/j.issn.1000-1298.2015.01.036.(in Chinese)
[30]陈锡龙.黄土高原地区高纺锤苹果树冠层3D模拟光截获与光合产物积累研究[D].杨凌:西北农林科技大学,2013.CHEN Xilong.An experimental study of light interception and photosynthesis product for 3D virtual tall-spindle apple canopy in Loess Plateau region of China[D].Yangling:Northwest A&F University,2013.(in Chinese)
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |