基于深度信息的大豆株高计算方法
|
摘要
为高通量地计算农作物株高,克服传统测量方法低效、耗时耗力等不足,以抗线9号、13号和富豆6号寒地大豆为研究对象,构建了基于Kinect 2.0的大豆冠层图像同步采集平台,并在三维重建大豆冠层结构形态的基础上,提出了基于深度信息的个体和群体大豆株高计算方法。实验结果表明,与实测值相比,计算得到的个体和群体大豆株高的平均误差分别为0.14cm和0.54cm,抗线9号、13号和富豆6号株高计算值与实测值之间的决定系数依次为0.9717,0.9730,0.9697。所提方法能够较为精确地计算大豆植株的株高特征。
In this study,three cold soybean varieties,namely Kangxian-9,Kangxian-13,and Fudou-6,were used for the high-throughput calculation of soybean plant heights.Three varieties were used to overcome the disadvantages of the traditional measurement methods,which were inefficient and laborious.The method for calculating the plant heights of individual and grouped soybean plants was proposed using the depth information acquired from the three-dimensional reconstruction of soybean canopies obtained using the Kinect V2.0 synchronous image acquisition platform.The experimental results show that compared with the measured value,the average errors of the proposed calculation method for the plant heights of individual and grouped soybean plants are 0.14 cm and 0.54 cm,respectively.The determination coefficients between calculated and measured values for Kangxian-9,Kangxian-13,and Fudou-6 are 0.9717,0.9730,and 0.9697,respectively.Thus,the proposed method can accurately calculate the heights of soybean plants.
In this study,three cold soybean varieties,namely Kangxian-9,Kangxian-13,and Fudou-6,were used for the high-throughput calculation of soybean plant heights.Three varieties were used to overcome the disadvantages of the traditional measurement methods,which were inefficient and laborious.The method for calculating the plant heights of individual and grouped soybean plants was proposed using the depth information acquired from the three-dimensional reconstruction of soybean canopies obtained using the Kinect V2.0 synchronous image acquisition platform.The experimental results show that compared with the measured value,the average errors of the proposed calculation method for the plant heights of individual and grouped soybean plants are 0.14 cm and 0.54 cm,respectively.The determination coefficients between calculated and measured values for Kangxian-9,Kangxian-13,and Fudou-6 are 0.9717,0.9730,and 0.9697,respectively.Thus,the proposed method can accurately calculate the heights of soybean plants.
引文
[1]Garrido M,Paraforos D S,Reiser D,et al.3D Maize plant reconstruction based on georeferenced overlapping LiDAR point clouds[J].Remote Sensing,2015,7(12):17077-17096.
[2]Lipka A E,Kandianis C B,Hudson M E,et al.From association to prediction:statistical methods for the dissection and selection of complex traits in plants[J].Current Opinion in Plant Biology,2015,24:110-118.
[3]Haughton A J,Bohan D A,Clark S J,et al.Dedicated biomass crops can enhance biodiversity in the arable landscape[J].Global Change Biology Bioenergy,2016,8(6):1071-1081.
[4]Song Q P,Tang J L,Xin J.3-dimensional reconstruction for soybean plant of seedling stage based on growth model[J].Computer Engineering,2017,43(5):275-280.宋祺鹏,唐晶磊,辛菁.基于生长模型的苗期大豆植株三维重建[J].计算机工程,2017,43(5):275-280.
[5]Pratap A,Tomar R,Kumar J,et al.High-throughput plant phenotyping platforms[J].Phenomics in Crop Plants:Trends Options and Limitations,2015:285-296.
[6]Barker III J,Zhang N Q,Sharon J,et al.Development of a field-based high-throughput mobile phenotyping platform[J].Computers and Electronics in Agriculture,2016,122:74-85.
[7]Palanichamy D,Cobb J N.Agronomic field trait phenomics[M].New York:Springer,2015:83-99.
[8]Li H.3D reconstruction of maize leaves based on virtual visual technology[J].Bulletin of Science and Technology,2016,32(5):96-101.李辉.基于虚拟双目视觉的玉米叶片三维重建方法[J].科技通报,2016,32(5):96-101.
[9]Li S H,Guan H O,Yu S,et al.Whole 3Dreconstruction based on fast SCAN corn and parameter calculation method[J].Journal of Agricultural Mechanization Research,2018,40(4):162-166.李抒昊,关海鸥,于崧,等.基于FastSCAN玉米整株三维重构及参数计算方法[J].农机化研究,2018,40(4):162-166.
[10]Jiang P,Geng N.Crop height non-destructive remote measurement based on IP camera[J].Journal of Agricultural Mechanization Research,2015,37(4):68-72.蒋普,耿楠.基于网络摄像机的株高远程无损测量系统[J].农机化研究,2015,37(4):68-72.
[11]Li Q,Cheng X J,Tian R,et al.Correction and normalization of multi-scan terrestrial three-dimensional laser scanning intensity[J].Laser&Optoelectronics Progress,2017,54(12):122802.李泉,程效军,田芮,等.多站地面三维激光扫描强度数据纠正与归一化[J].激光与光电子学进展,2017,54(12):122802.
[12]Xiao Y,Hu S X,Xiao S,et al.A fast statistical method of tree information from 3D laser point clouds[J].Chinese Journal of Lasers,2018,45(5):0510007.肖杨,胡少兴,肖深,等.从三维激光点云中快速统计树木信息的方法[J].中国激光,2018,45(5):0510007.
[13]Ma X D,Guo C L,Zhang X,et al.Calculation of light distribution of apple tree canopy based on color characteristics of 3Dpoint cloud[J].Transactions of the Chinese Society for Agricultural Machinery,2015,46(6):263-268.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].农业机械学报,2015,46(6):263-268.
[14]Ma X D,Guo C L,Zhang X,et al.Calculation of chlorophyll fluorescence characters in different light area to apple tree canopy[J].Spectroscopy and Spectral Analysis,2016,36(12):3986-3990.马晓丹,郭彩玲,张雪,等.苹果树冠层不同光照区域叶绿素荧光性状计算方法[J].光谱学与光谱分析,2016,36(12):3986-3990.
[15]Jiang Y,Li C Y,Paterson A H.High throughput phenotyping of cotton plant height using depth images under field conditions[J].Computers and Electronics in Agriculture,2016,130:57-68.
[16]Ma X D,Feng J R,Guan H O,et al.Prediction of chlorophyll content in different light areas of apple tree canopies based on the color characteristics of 3Dreconstruction[J].Remote Sensing,2018,10(3):429.
[17]Skerik F,Volf J,Linda M,et al.Methods for detection of objects in agriculture,Prague[C]∥Conference Proceeding-5th International Conference,TAE 2013:Trends in Agricultural Engineering,September 3-6,2013.Czech University of Life Sciences Prague,2013,599-601.
[18]Yamamoto S,Hayashi S,Saito S,et al.Measurement of growth information of a strawberry plantusing a natural interaction device[C]∥2012Dallas,Texas,July 29-August 1,2012.American Society of Agricultural and Biological Engineers,2012:121341108.
[19]Shen Y,Zhu J H,Liu H,et al.Plant image mosaic based on depth and color dual information feature source from Kinect[J].Transactions of the Chinese Society of Agricultural Engineering,2018,34(5):176-182.沈跃,朱嘉慧,刘慧,等.基于深度和彩色双信息特征源的Kinect植物图像拼接[J].农业工程学报,2018,34(5):176-182.
[20]Liu M,Bu W Z,Yang W Y,et al.Correlation analysis of yield and agronomic traits of soybean for intercropping in Shangdong[J].Chinese Journal of Oil Crop Sciences,2018,40(3):344-351.刘明,卜伟召,杨文钰,等.山东间作大豆产量与主要农艺性状关联分析[J].中国油料作物学报,2018,40(3):344-351.
[21]Liu G,Zhang X,Zong Z,et al.3Dreconstruction of strawberry based on depth information[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(4):160-165,172.刘刚,张雪,宗泽,等.基于深度信息的草莓三维重建技术[J].农业机械学报,2017,48(4):160-165,172.
[22]Yu X L,Wang D D,Niu L L,et al.The applications and research progress of kinect in the information field of modern agriculturae[J].Journal of Agricultural Mechanization Research,2015,37(11):216-221.余秀丽,王丹丹,牛磊磊,等.Kinect在现代农业信息领域中的应用与研究进展[J].农机化研究,2015,37(11):216-221.
[23]Guo L P,Chen X N,Liu B.Calibration of Kinect sensor with depth and color camera[J].Journal of Image and Graphics,2014,19(11):1584-1590.郭连朋,陈向宁,刘彬.Kinect传感器的彩色和深度相机标定[J].中国图象图形学报,2014,19(11):1584-1590.
[24]Li Y N.Research on calibration algorithm of Kinect depth camera[D].Xi′an:Xidian University,2014.李雅娜.Kinect深度相机标定算法研究[D].西安:西安电子科技大学,2014.
[25]Wang S,Xu X.3Dreconstruction based on horopter[J].Acta Optica Sinica,2017,37(5):0515004.王珊,徐晓.基于双目单视面的三维重建[J].光学学报,2017,37(5):0515004.
[26]Zhou W,Ma X D,Zhang L J,et al.Three dimensional point cloud splicing of tree canopy based on multi-source camera[J].Acta Optica Sinica,2014,34(12):1215003.周薇,马晓丹,张丽娇,等.基于多源信息融合的果树冠层三维点云拼接方法研究[J].光学学报,2014,34(12):1215003.
[27]Wang M B.Study on three-dimensional reconstruction of crop plants based on point cloud[J].Journal of Ezhou University,2018,25(1):104-106,109.王孟博.基于点云的作物植株三维重建技术研究[J].鄂州大学学报,2018,25(1):104-106,109.
[28]Li X Q.Research on image threshold segmentation algorithm based on Matlab[J].Software Guide,2014,13(12):76-78.李小琦.基于Matlab的图像阈值分割算法研究[J].软件导刊,2014,13(12):76-78.
[29]Guan H O,Liu M,Ma X D,et al.Threedimensional reconstruction of soybean canopies using multisource imaging for phenotyping analysis[J].Remote Sensing,2018,10(8):1206.
[30]Zhu B L,Liu F S,Zhu J Y,et al.Three-dimensional quantifications of plant growth dynamics in fieldgrown plants based on machine vision method[J].Transactions of the Chinese Society for Agricultural Machinery,2018,49(5):256-262.朱冰琳,刘扶桑,朱晋宇,等.基于机器视觉的大田植株生长动态三维定量化研究[J].农业机械学报,2018,49(5):256-262.
[31]Carlone L,Dong J,Fenu S,et al.Towards 4Dcrop analysis in precision agriculture:estimating plant height and crown radius over time via expectationmaximization[C]∥IEEE ICRA Workshop on Robotics in Agriculture,May 26-30,2015Seattle,WA,USA,2015.
[2]Lipka A E,Kandianis C B,Hudson M E,et al.From association to prediction:statistical methods for the dissection and selection of complex traits in plants[J].Current Opinion in Plant Biology,2015,24:110-118.
[3]Haughton A J,Bohan D A,Clark S J,et al.Dedicated biomass crops can enhance biodiversity in the arable landscape[J].Global Change Biology Bioenergy,2016,8(6):1071-1081.
[4]Song Q P,Tang J L,Xin J.3-dimensional reconstruction for soybean plant of seedling stage based on growth model[J].Computer Engineering,2017,43(5):275-280.宋祺鹏,唐晶磊,辛菁.基于生长模型的苗期大豆植株三维重建[J].计算机工程,2017,43(5):275-280.
[5]Pratap A,Tomar R,Kumar J,et al.High-throughput plant phenotyping platforms[J].Phenomics in Crop Plants:Trends Options and Limitations,2015:285-296.
[6]Barker III J,Zhang N Q,Sharon J,et al.Development of a field-based high-throughput mobile phenotyping platform[J].Computers and Electronics in Agriculture,2016,122:74-85.
[7]Palanichamy D,Cobb J N.Agronomic field trait phenomics[M].New York:Springer,2015:83-99.
[8]Li H.3D reconstruction of maize leaves based on virtual visual technology[J].Bulletin of Science and Technology,2016,32(5):96-101.李辉.基于虚拟双目视觉的玉米叶片三维重建方法[J].科技通报,2016,32(5):96-101.
[9]Li S H,Guan H O,Yu S,et al.Whole 3Dreconstruction based on fast SCAN corn and parameter calculation method[J].Journal of Agricultural Mechanization Research,2018,40(4):162-166.李抒昊,关海鸥,于崧,等.基于FastSCAN玉米整株三维重构及参数计算方法[J].农机化研究,2018,40(4):162-166.
[10]Jiang P,Geng N.Crop height non-destructive remote measurement based on IP camera[J].Journal of Agricultural Mechanization Research,2015,37(4):68-72.蒋普,耿楠.基于网络摄像机的株高远程无损测量系统[J].农机化研究,2015,37(4):68-72.
[11]Li Q,Cheng X J,Tian R,et al.Correction and normalization of multi-scan terrestrial three-dimensional laser scanning intensity[J].Laser&Optoelectronics Progress,2017,54(12):122802.李泉,程效军,田芮,等.多站地面三维激光扫描强度数据纠正与归一化[J].激光与光电子学进展,2017,54(12):122802.
[12]Xiao Y,Hu S X,Xiao S,et al.A fast statistical method of tree information from 3D laser point clouds[J].Chinese Journal of Lasers,2018,45(5):0510007.肖杨,胡少兴,肖深,等.从三维激光点云中快速统计树木信息的方法[J].中国激光,2018,45(5):0510007.
[13]Ma X D,Guo C L,Zhang X,et al.Calculation of light distribution of apple tree canopy based on color characteristics of 3Dpoint cloud[J].Transactions of the Chinese Society for Agricultural Machinery,2015,46(6):263-268.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].农业机械学报,2015,46(6):263-268.
[14]Ma X D,Guo C L,Zhang X,et al.Calculation of chlorophyll fluorescence characters in different light area to apple tree canopy[J].Spectroscopy and Spectral Analysis,2016,36(12):3986-3990.马晓丹,郭彩玲,张雪,等.苹果树冠层不同光照区域叶绿素荧光性状计算方法[J].光谱学与光谱分析,2016,36(12):3986-3990.
[15]Jiang Y,Li C Y,Paterson A H.High throughput phenotyping of cotton plant height using depth images under field conditions[J].Computers and Electronics in Agriculture,2016,130:57-68.
[16]Ma X D,Feng J R,Guan H O,et al.Prediction of chlorophyll content in different light areas of apple tree canopies based on the color characteristics of 3Dreconstruction[J].Remote Sensing,2018,10(3):429.
[17]Skerik F,Volf J,Linda M,et al.Methods for detection of objects in agriculture,Prague[C]∥Conference Proceeding-5th International Conference,TAE 2013:Trends in Agricultural Engineering,September 3-6,2013.Czech University of Life Sciences Prague,2013,599-601.
[18]Yamamoto S,Hayashi S,Saito S,et al.Measurement of growth information of a strawberry plantusing a natural interaction device[C]∥2012Dallas,Texas,July 29-August 1,2012.American Society of Agricultural and Biological Engineers,2012:121341108.
[19]Shen Y,Zhu J H,Liu H,et al.Plant image mosaic based on depth and color dual information feature source from Kinect[J].Transactions of the Chinese Society of Agricultural Engineering,2018,34(5):176-182.沈跃,朱嘉慧,刘慧,等.基于深度和彩色双信息特征源的Kinect植物图像拼接[J].农业工程学报,2018,34(5):176-182.
[20]Liu M,Bu W Z,Yang W Y,et al.Correlation analysis of yield and agronomic traits of soybean for intercropping in Shangdong[J].Chinese Journal of Oil Crop Sciences,2018,40(3):344-351.刘明,卜伟召,杨文钰,等.山东间作大豆产量与主要农艺性状关联分析[J].中国油料作物学报,2018,40(3):344-351.
[21]Liu G,Zhang X,Zong Z,et al.3Dreconstruction of strawberry based on depth information[J].Transactions of the Chinese Society for Agricultural Machinery,2017,48(4):160-165,172.刘刚,张雪,宗泽,等.基于深度信息的草莓三维重建技术[J].农业机械学报,2017,48(4):160-165,172.
[22]Yu X L,Wang D D,Niu L L,et al.The applications and research progress of kinect in the information field of modern agriculturae[J].Journal of Agricultural Mechanization Research,2015,37(11):216-221.余秀丽,王丹丹,牛磊磊,等.Kinect在现代农业信息领域中的应用与研究进展[J].农机化研究,2015,37(11):216-221.
[23]Guo L P,Chen X N,Liu B.Calibration of Kinect sensor with depth and color camera[J].Journal of Image and Graphics,2014,19(11):1584-1590.郭连朋,陈向宁,刘彬.Kinect传感器的彩色和深度相机标定[J].中国图象图形学报,2014,19(11):1584-1590.
[24]Li Y N.Research on calibration algorithm of Kinect depth camera[D].Xi′an:Xidian University,2014.李雅娜.Kinect深度相机标定算法研究[D].西安:西安电子科技大学,2014.
[25]Wang S,Xu X.3Dreconstruction based on horopter[J].Acta Optica Sinica,2017,37(5):0515004.王珊,徐晓.基于双目单视面的三维重建[J].光学学报,2017,37(5):0515004.
[26]Zhou W,Ma X D,Zhang L J,et al.Three dimensional point cloud splicing of tree canopy based on multi-source camera[J].Acta Optica Sinica,2014,34(12):1215003.周薇,马晓丹,张丽娇,等.基于多源信息融合的果树冠层三维点云拼接方法研究[J].光学学报,2014,34(12):1215003.
[27]Wang M B.Study on three-dimensional reconstruction of crop plants based on point cloud[J].Journal of Ezhou University,2018,25(1):104-106,109.王孟博.基于点云的作物植株三维重建技术研究[J].鄂州大学学报,2018,25(1):104-106,109.
[28]Li X Q.Research on image threshold segmentation algorithm based on Matlab[J].Software Guide,2014,13(12):76-78.李小琦.基于Matlab的图像阈值分割算法研究[J].软件导刊,2014,13(12):76-78.
[29]Guan H O,Liu M,Ma X D,et al.Threedimensional reconstruction of soybean canopies using multisource imaging for phenotyping analysis[J].Remote Sensing,2018,10(8):1206.
[30]Zhu B L,Liu F S,Zhu J Y,et al.Three-dimensional quantifications of plant growth dynamics in fieldgrown plants based on machine vision method[J].Transactions of the Chinese Society for Agricultural Machinery,2018,49(5):256-262.朱冰琳,刘扶桑,朱晋宇,等.基于机器视觉的大田植株生长动态三维定量化研究[J].农业机械学报,2018,49(5):256-262.
[31]Carlone L,Dong J,Fenu S,et al.Towards 4Dcrop analysis in precision agriculture:estimating plant height and crown radius over time via expectationmaximization[C]∥IEEE ICRA Workshop on Robotics in Agriculture,May 26-30,2015Seattle,WA,USA,2015.