基于形态重构的叶片性状特征可视化表达方法
|
摘要
针对目前叶片性状特征在信息融合与表达过程中存在单一性及抽象性等问题,提出一种基于形态重构的叶片性状特征可视化表达方法。以温室黄瓜叶片生长为例,将有效积温、生长速率等作为特征参数,建立叶片形态发生模型,利用参数化样条曲线描述叶缘、叶脉的几何形态,采取二分法递归地分割叶缘及叶脉曲线,以实现叶片曲面的网格化细分,结合叶色纹理信息映射模型,提出叶片性状特征的可视化表达方法。实验结果表明,运用该方法得到的叶片性状特征模拟值与观测值之间的相对误差较小,其决定系数均达到0. 95以上,均方根误差不大于0. 236,与传统的建模方法相比,该模型具有更高的拟合度和可靠性,能够有效实现黄瓜叶片性状变化的动态仿真,可为实时掌握和预测植物生长发育状况提供依据。
Leaf traits can provide important references for canopy light distribution, growth and development,and monitoring of external environment. Aiming at the problems of simplicity and abstraction in the process of processing and expressing leaf traits,a leaf traits fusion method based on morphological reconstruction was proposed. Taking the growth of cucumber leaves in greenhouse as an example,the effective accumulated temperature and growth rate were taken as characteristic parameters to establish the leaf morphogenesis model. The parametric spline curve was used to describe the geometric shape of leaf edge and vein. The dichotomy method was used to divide the leaf edge and vein curve recursively in order to realize the meshed subdivision of the blade surface. Combining with the leaf color texture information mapping model,a visual expression algorithm of leaf characteristics was introduced.The experimental verification results showed that the relative errors between the simulated and observed values of leaf traits obtained by this method were small, and the consistency was good, which demonstrated the method had certain feasibility and validity. Furthermore,in comparison with the typical statistical model and point cloud reconstruction model,the experimental results indicated that the square of correlation coefficients was above 0. 95,and the root mean square deviation was no more than 0. 236.Compared with traditional modeling methods,the proposed model had higher fitting degree and better reliability,by which it can effectively realize the dynamic simulation of cucumber leaf traits,which could provide a basis for real-time grasping and forecasting of plant growth and development. This method not only provided a reference for the dynamic tracking and management of greenhouse crop production,but also laid a theoretical foundation for the further study of the role of plants under various environmental factors.
Leaf traits can provide important references for canopy light distribution, growth and development,and monitoring of external environment. Aiming at the problems of simplicity and abstraction in the process of processing and expressing leaf traits,a leaf traits fusion method based on morphological reconstruction was proposed. Taking the growth of cucumber leaves in greenhouse as an example,the effective accumulated temperature and growth rate were taken as characteristic parameters to establish the leaf morphogenesis model. The parametric spline curve was used to describe the geometric shape of leaf edge and vein. The dichotomy method was used to divide the leaf edge and vein curve recursively in order to realize the meshed subdivision of the blade surface. Combining with the leaf color texture information mapping model,a visual expression algorithm of leaf characteristics was introduced.The experimental verification results showed that the relative errors between the simulated and observed values of leaf traits obtained by this method were small, and the consistency was good, which demonstrated the method had certain feasibility and validity. Furthermore,in comparison with the typical statistical model and point cloud reconstruction model,the experimental results indicated that the square of correlation coefficients was above 0. 95,and the root mean square deviation was no more than 0. 236.Compared with traditional modeling methods,the proposed model had higher fitting degree and better reliability,by which it can effectively realize the dynamic simulation of cucumber leaf traits,which could provide a basis for real-time grasping and forecasting of plant growth and development. This method not only provided a reference for the dynamic tracking and management of greenhouse crop production,but also laid a theoretical foundation for the further study of the role of plants under various environmental factors.
引文
[1] HU B G,ZHAO X,YAN H P,et al. Plant growth modeling and visualization-review and perspective[J]. Acta Automatica Sinica,2001,27(4):567-584.
[2]杨其长,魏灵玲,刘文科,等.中国设施农业研究现状及发展战略[J].中国农业信息,2012(11):22-27.YANG Qichang,WEI Lingling,LIU Wenke,et al. Research and development strategy of facility agriculture in China[J]. China Agriculture Information,2012(11):22-27.(in Chinese)
[3]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)
[4] KOLECKA N,KOZAK J,KAIM D,et al. Mapping secondary forest succession on abandoned agricultural land with Li DAR point clouds and terrestrial photography[J]. Remote Sensing,2015,7(7):8300-8322.
[5]吴露露,马旭,齐龙,等.基于叶片形态的田间植物检测方法[J/OL].农业机械学报,2013,44(11):241-246.WU Lulu,MA Xu,QI Long,et al. Field plants detection based on leaf morphology[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2013,44(11):241-246. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20131141&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2013. 11. 041.(in Chinese)
[6] MARTINEZ J A,RUFAT J,ARNO J,et al. Mobile terrestrial laser scanner applications in precision fruticulture/horticulture and tools to extract information from canopy point clouds[J]. Precision Agriculture,2017,18(1):111-132.
[7]姜文涛,刘万军,袁姮.基于软特征理论的目标跟踪研究[J].计算机学报,2016,39(7):1334-1355.JIANG Wentao,LIU Wanjun,YUAN Heng. Research of object tracking based on soft feature theory[J]. Chinese Journal of Computers,2016,39(7):1334-1355.(in Chinese)
[8] MEDEIROS H,KIM D,SUN J,et al. Modeling dormant fruit trees for agricultural automation[J]. Journal of Field Robotics,2017,34(7):1203-1224.
[9] LEHNERT C,ENGLISH A,MCCOOL C,et al. Autonomous sweet pepper harvesting for protected cropping systems[J]. IEEE Robotics and Automation Letters,2017,2(2):872-879.
[10]杨亮,郭新宇,陆声链,等.基于多幅图像的黄瓜叶片形态三维重建[J].农业工程学报,2009,25(2):141-144.YANG Liang,GUO Xinyu,LU Shenglian,et al. 3D morphological reconstruction of cucumber leaf based on multiple images[J]. Transactions of the CSAE,2009,25(2):141-144.(in Chinese)
[11]王树文,张长利.基于图像处理技术的黄瓜叶片病害识别诊断系统研究[J].东北农业大学学报,2012,43(5):69-73.WANG Shuwen,ZHANG Changli. Study on identification of cucumber leaf disease based on image processing[J]. Journal of Northeast Agricultural University,2012,43(5):69-73.(in Chinese)
[12] LI Z,ISLER V. Large scale image mosaic construction for agricultural applications[J]. IEEE Robotics&Automation Letters,2016,1(1):295-302.
[13]乔虹,冯全,张芮,等.基于时序图像跟踪的葡萄叶片病害动态监测[J].农业工程学报,2018,34(17):167-175.QIAO Hong,FENG Quan,ZHANG Rui,et al. Dynamic monitoring of grape leaf disease based on sequential images tracking[J]. Transactions of the CSAE,2018,34(17):167-175.(in Chinese)
[14]刘媛.基于深度学习的葡萄叶片病害识别方法研究[D].兰州:甘肃农业大学,2018.LIU Yuan. Research on methods on grape leaf disease recognition based on deep learning[D]. Lanzhou:Gansu Agricultural University,2018.(in Chinese)
[15]孙智慧,陆声链,郭新宇,等.基于点云数据的植物叶片曲面重构方法[J].农业工程学报,2012,28(3):184-190.SUN Zhihui,LU Shenglian,GUO Xinyu,et al. Surfaces reconstruction of plant leaves based on point cloud data[J].Transactions of the CSAE,2012,28(3):184-190.(in Chinese)
[16]王志彬,王开义,王书锋,等.基于动态集成的黄瓜叶部病害识别方法[J/OL].农业机械学报,2017,48(9):46-52.WANG Zhibin,WANG Kaiyi,WANG Shufeng,et al. Recognition method of cucumber leaf diseases with dynamic ensemble learning[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(9):46-52. http:∥www. j-csam.org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170906&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2017. 09. 006.(in Chinese)
[17] 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.
[18] JORDAN U,MIKOLAJ C,PRZEMYSLAW P,et al. The use of plant models in deep learning:an application to leaf counting in rosette plants[J]. Plant Methods,2018,14:6-15.
[19] KEITA M,MIKI O,CHIAKI Y,et al. Gravitropism interferes with hydrotropism via counteracting auxin dynamics in cucumber roots:clinorotation and spaceflight experiments[J]. New Phytologist,2017,215:1476-1489.
[20]蔡祥,孙宇瑞,赵燕东,等.基于3-D数据的叶片萎蔫形态辨识方法[J/OL].农业机械学报,2015,46(5):286-292.CAI Xiang,SUN Yurui,ZHAO Yandong,et al. Identification methods for plant wilting at leaf-scale based on 3-D imaging[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2015,46(5):286-292. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20150541&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2015. 05.041.(in Chinese)
[21] VO A V,LINH T H,LAEFER D F,et al. Octree-based region growing for point cloud segmentation[J]. ISPRS Journal of Photogrammetry&Remote Sensing,2015,104:88-100.
[22] HAMUDA E,MC G B,GLAVIN M,et al. Automatic crop detection under field conditions using the HSV colour space and morphological operations[J]. Computers and Electronics in Agriculture,2017,133:97-107.
[23]徐艳蕾,贾洪雷,包佳林.基于模糊顺序形态学的植物叶片脉络边缘提取[J].农业工程学报,2015,31(13):193-198.XU Yanlei,JIA Honglei,BAO Jialin. Plant leaf vein edge detection based on fuzzy order morphology[J]. Transactions of the CSAE,2015,31(13):193-198.(in Chinese)
[24]肖伯祥,郭新宇,王纪华,等.玉米叶片形态建模与网格简化算法研究[J].中国农业科学,2007,40(4):693-697.XIAO Boxiang,GUO Xinyu,WANG Jihua,et al. Maize leaf morphological modeling and mesh simplification of surface[J].Scientia Agricultura Sinica,2007,40(4):693-697.(in Chinese)
[25]苗腾,赵春江,郭新宇,等.基于数据约束的黄瓜叶片参数化建模[J].农业机械学报,2011,42(1):171-176.MIAO Teng,ZHAO Chunjiang,GUO Xinyu,et al. Parameterized geometric modeling of cucumber leaf based on data constrains[J]. Transactions of the Chinese Society for Agricultural Machinery,2011,42(1):171-176.(in Chinese)
[26] SYLVAIN D,CHRISTIAN J,PASCAL R. Pypetree:a tool for reconstructing tree perennial tissues from point clouds[J].Sensors,2014,14(3):4271-4289.
[27]吴升,赵春江,郭新宇,等.基于点云的果树冠层叶片重建方法[J].农业工程学报,2017,33(增刊1):212-218.WU Sheng,ZHAO Chunjiang,GUO Xinyu,et al. Method of fruit tree canopy leaf reconstruction based on point cloud[J].Transactions of the CSAE,2017,33(Supp. 1):212-218.(in Chinese)
[28]杨军,林岩龙,李龙杰,等.基于快速Delaunay三角化的散乱点曲面重建算法[J].计算机工程与科学,2015,37(6):1189-1195.YANG Jun,LIN Yanlong,LI Longjie,et al. A surface reconstruction algorithm for unorganized points based on fast Delaunay triangulation[J]. Computer Engineering&Science,2015,37(6):1189-1195.(in Chinese)
[29]郑一力,钟刚亮,王强,等.基于多特征降维的植物叶片识别方法[J/OL].农业机械学报,2017,48(3):30-37.ZHENG Yili,ZHONG Gangliang,WANG Qiang,et al. Method of leaf identification based on multi-feature dimension reduction[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):30-37. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170304&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2017. 03. 004.(in Chinese)
[30] LIU L,FIEGUTH P,GUO Y,et al. Local binary features for texture classification:taxonomy and experimental study[J].Pattern Recognition,2017,62:135-160.
[31]邓向武,齐龙,马旭,等.基于多特征融合和深度置信网络的稻田苗期杂草识别[J].农业工程学报,2018,34(14):165-172.DENG Xiangwu,QI Long,MA Xu,et al. Recognition of weeds at seedling stage in paddy fields using multi-feature fusion and deep belief networks[J]. Transactions of the CSAE,2018,34(14):165-172.(in Chinese)
[2]杨其长,魏灵玲,刘文科,等.中国设施农业研究现状及发展战略[J].中国农业信息,2012(11):22-27.YANG Qichang,WEI Lingling,LIU Wenke,et al. Research and development strategy of facility agriculture in China[J]. China Agriculture Information,2012(11):22-27.(in Chinese)
[3]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)
[4] KOLECKA N,KOZAK J,KAIM D,et al. Mapping secondary forest succession on abandoned agricultural land with Li DAR point clouds and terrestrial photography[J]. Remote Sensing,2015,7(7):8300-8322.
[5]吴露露,马旭,齐龙,等.基于叶片形态的田间植物检测方法[J/OL].农业机械学报,2013,44(11):241-246.WU Lulu,MA Xu,QI Long,et al. Field plants detection based on leaf morphology[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2013,44(11):241-246. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20131141&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2013. 11. 041.(in Chinese)
[6] MARTINEZ J A,RUFAT J,ARNO J,et al. Mobile terrestrial laser scanner applications in precision fruticulture/horticulture and tools to extract information from canopy point clouds[J]. Precision Agriculture,2017,18(1):111-132.
[7]姜文涛,刘万军,袁姮.基于软特征理论的目标跟踪研究[J].计算机学报,2016,39(7):1334-1355.JIANG Wentao,LIU Wanjun,YUAN Heng. Research of object tracking based on soft feature theory[J]. Chinese Journal of Computers,2016,39(7):1334-1355.(in Chinese)
[8] MEDEIROS H,KIM D,SUN J,et al. Modeling dormant fruit trees for agricultural automation[J]. Journal of Field Robotics,2017,34(7):1203-1224.
[9] LEHNERT C,ENGLISH A,MCCOOL C,et al. Autonomous sweet pepper harvesting for protected cropping systems[J]. IEEE Robotics and Automation Letters,2017,2(2):872-879.
[10]杨亮,郭新宇,陆声链,等.基于多幅图像的黄瓜叶片形态三维重建[J].农业工程学报,2009,25(2):141-144.YANG Liang,GUO Xinyu,LU Shenglian,et al. 3D morphological reconstruction of cucumber leaf based on multiple images[J]. Transactions of the CSAE,2009,25(2):141-144.(in Chinese)
[11]王树文,张长利.基于图像处理技术的黄瓜叶片病害识别诊断系统研究[J].东北农业大学学报,2012,43(5):69-73.WANG Shuwen,ZHANG Changli. Study on identification of cucumber leaf disease based on image processing[J]. Journal of Northeast Agricultural University,2012,43(5):69-73.(in Chinese)
[12] LI Z,ISLER V. Large scale image mosaic construction for agricultural applications[J]. IEEE Robotics&Automation Letters,2016,1(1):295-302.
[13]乔虹,冯全,张芮,等.基于时序图像跟踪的葡萄叶片病害动态监测[J].农业工程学报,2018,34(17):167-175.QIAO Hong,FENG Quan,ZHANG Rui,et al. Dynamic monitoring of grape leaf disease based on sequential images tracking[J]. Transactions of the CSAE,2018,34(17):167-175.(in Chinese)
[14]刘媛.基于深度学习的葡萄叶片病害识别方法研究[D].兰州:甘肃农业大学,2018.LIU Yuan. Research on methods on grape leaf disease recognition based on deep learning[D]. Lanzhou:Gansu Agricultural University,2018.(in Chinese)
[15]孙智慧,陆声链,郭新宇,等.基于点云数据的植物叶片曲面重构方法[J].农业工程学报,2012,28(3):184-190.SUN Zhihui,LU Shenglian,GUO Xinyu,et al. Surfaces reconstruction of plant leaves based on point cloud data[J].Transactions of the CSAE,2012,28(3):184-190.(in Chinese)
[16]王志彬,王开义,王书锋,等.基于动态集成的黄瓜叶部病害识别方法[J/OL].农业机械学报,2017,48(9):46-52.WANG Zhibin,WANG Kaiyi,WANG Shufeng,et al. Recognition method of cucumber leaf diseases with dynamic ensemble learning[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(9):46-52. http:∥www. j-csam.org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170906&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2017. 09. 006.(in Chinese)
[17] 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.
[18] JORDAN U,MIKOLAJ C,PRZEMYSLAW P,et al. The use of plant models in deep learning:an application to leaf counting in rosette plants[J]. Plant Methods,2018,14:6-15.
[19] KEITA M,MIKI O,CHIAKI Y,et al. Gravitropism interferes with hydrotropism via counteracting auxin dynamics in cucumber roots:clinorotation and spaceflight experiments[J]. New Phytologist,2017,215:1476-1489.
[20]蔡祥,孙宇瑞,赵燕东,等.基于3-D数据的叶片萎蔫形态辨识方法[J/OL].农业机械学报,2015,46(5):286-292.CAI Xiang,SUN Yurui,ZHAO Yandong,et al. Identification methods for plant wilting at leaf-scale based on 3-D imaging[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2015,46(5):286-292. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20150541&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298. 2015. 05.041.(in Chinese)
[21] VO A V,LINH T H,LAEFER D F,et al. Octree-based region growing for point cloud segmentation[J]. ISPRS Journal of Photogrammetry&Remote Sensing,2015,104:88-100.
[22] HAMUDA E,MC G B,GLAVIN M,et al. Automatic crop detection under field conditions using the HSV colour space and morphological operations[J]. Computers and Electronics in Agriculture,2017,133:97-107.
[23]徐艳蕾,贾洪雷,包佳林.基于模糊顺序形态学的植物叶片脉络边缘提取[J].农业工程学报,2015,31(13):193-198.XU Yanlei,JIA Honglei,BAO Jialin. Plant leaf vein edge detection based on fuzzy order morphology[J]. Transactions of the CSAE,2015,31(13):193-198.(in Chinese)
[24]肖伯祥,郭新宇,王纪华,等.玉米叶片形态建模与网格简化算法研究[J].中国农业科学,2007,40(4):693-697.XIAO Boxiang,GUO Xinyu,WANG Jihua,et al. Maize leaf morphological modeling and mesh simplification of surface[J].Scientia Agricultura Sinica,2007,40(4):693-697.(in Chinese)
[25]苗腾,赵春江,郭新宇,等.基于数据约束的黄瓜叶片参数化建模[J].农业机械学报,2011,42(1):171-176.MIAO Teng,ZHAO Chunjiang,GUO Xinyu,et al. Parameterized geometric modeling of cucumber leaf based on data constrains[J]. Transactions of the Chinese Society for Agricultural Machinery,2011,42(1):171-176.(in Chinese)
[26] SYLVAIN D,CHRISTIAN J,PASCAL R. Pypetree:a tool for reconstructing tree perennial tissues from point clouds[J].Sensors,2014,14(3):4271-4289.
[27]吴升,赵春江,郭新宇,等.基于点云的果树冠层叶片重建方法[J].农业工程学报,2017,33(增刊1):212-218.WU Sheng,ZHAO Chunjiang,GUO Xinyu,et al. Method of fruit tree canopy leaf reconstruction based on point cloud[J].Transactions of the CSAE,2017,33(Supp. 1):212-218.(in Chinese)
[28]杨军,林岩龙,李龙杰,等.基于快速Delaunay三角化的散乱点曲面重建算法[J].计算机工程与科学,2015,37(6):1189-1195.YANG Jun,LIN Yanlong,LI Longjie,et al. A surface reconstruction algorithm for unorganized points based on fast Delaunay triangulation[J]. Computer Engineering&Science,2015,37(6):1189-1195.(in Chinese)
[29]郑一力,钟刚亮,王强,等.基于多特征降维的植物叶片识别方法[J/OL].农业机械学报,2017,48(3):30-37.ZHENG Yili,ZHONG Gangliang,WANG Qiang,et al. Method of leaf identification based on multi-feature dimension reduction[J/OL]. Transactions of the Chinese Society for Agricultural Machinery,2017,48(3):30-37. http:∥www. j-csam. org/jcsam/ch/reader/view_abstract. aspx? flag=1&file_no=20170304&journal_id=jcsam. DOI:10. 6041/j. issn. 1000-1298.2017. 03. 004.(in Chinese)
[30] LIU L,FIEGUTH P,GUO Y,et al. Local binary features for texture classification:taxonomy and experimental study[J].Pattern Recognition,2017,62:135-160.
[31]邓向武,齐龙,马旭,等.基于多特征融合和深度置信网络的稻田苗期杂草识别[J].农业工程学报,2018,34(14):165-172.DENG Xiangwu,QI Long,MA Xu,et al. Recognition of weeds at seedling stage in paddy fields using multi-feature fusion and deep belief networks[J]. Transactions of the CSAE,2018,34(14):165-172.(in Chinese)