基于有效积温的冬小麦返青后植株三维形态模拟  

3D Shape Simulation of Winter Wheat after Turning Green Stage Based on Effective Accumulated Temperature

作　　者：李书钦[1,2] 诸叶平[1] 刘海龙[1] 李世娟[1] 刘升平[1] 张红英[1] 高伟[3] 

机构地区：[1]中国农业科学院农业信息研究所/农业部农业信息服务技术重点实验室,北京100081 [2]北方工业大学信息中心,北京100144 [3]天津市农业科学院农业资源与环境研究所,天津300192

出　　处：《中国农业科学》2017年第9期1594-1605,共12页Scientia Agricultura Sinica

基　　金：国家"863"计划项目(2013AA102305);国家自然科学基金(61271364);国家重点研发计划项目(2016YFD0200601);中国农业科学院科技创新工程项目(CAAS-ASTIP-2016-AII-03);中国农业科学院协同创新项目(CAAS-XTCX2016006)

摘　　要：【目的】基于有效积温,利用三维建模技术,实现小麦生长模型与形态模型的有机结合,真实表达环境因素对小麦生长发育和形态结构的影响,最终实现小麦生长过程的三维可视化,为小麦作物生长动态预测、栽培管理调控、作物株型设计等提供重要参考。【方法】以天津地区主要推广小麦品种衡观35、济麦22和衡4399为材料,于2015—2016年冬小麦生长季内开展不同小麦品种和施氮水平的田间试验,采集各品种冬小麦在不同施氮水平下的叶长和最大叶宽等形态数据,通过分析各品种冬小麦返青后形态数据和有效积温的定量关系,用Logistic方程构建了冬小麦返青后叶片叶长、最大叶宽模拟模型,并对该模型进行检验;基于该模拟模型,计算各品种冬小麦返青后每个生长日的形态数据,借助OpenGL和NURBS曲面造型技术,构建冬小麦几何形态模型,最终实现冬小麦生长模型与形态模型的结合,实现了冬小麦返青后生长过程可视化。【结果】在不同品种、不同施氮水平下,小麦叶长回归方程R2值在0.772—0.983之间,F值在10.153—340.191之间,且Sig小于显著水平0.05,最大叶宽回归方程R^2值在0.853—0.999之间,F值在17.371—4 359.236之间,且Sig小于显著水平0.05,表明上述模型拟合度和显著性均较好。经数据检验,叶长模型绝对误差在0—3.88 cm之间,根均方差(RMSE)值在0.24—1.95 cm之间,最大叶宽模型绝对误差在0—0.28 cm之间,RMSE值在0.02—0.15 cm之间,表明所建模拟模型精度较高,该模型对不同品种冬小麦返青后的叶片生长具有较好的预测性;基于所建模拟模型计算冬小麦返青后逐日形态数据,可构造不同品种、不同施氮水平下的冬小麦植株形态,可逼真模拟冬小麦返青后植株动态生长过程。【结论】基于有效积温构建的冬小麦返青后叶长和最大叶宽模拟模型,可较好预测冬小麦返青后叶片生长状态,可实现�[Objective] Based on effective accumulated temperature, the aim of this study is to realize combination of wheat growth model and shape model using 3D modeling technology, express environmental factors influence on wheat growth and morphological structure, finally realize the 3D visualization in the process of wheat growth, provide important reference for wheat crop growth dynamic prediction, cultivation management control and crop plant type design. [Method] As the main commercial wheat varieties in Tianjin region, Hengguan35, Jimai22 and Heng4399 were used as the experimental materials in this study, the field experiments of different varieties and nitrogen levels were carried out in 2015-2016 growth seasons of winter wheat, winter wheat shape data were collected under different nitrogen levels. After analysis of quantitative relationship among various varieties of winter wheat morphology data and effective accumulated temperature, simulation models of winter wheat leaf length and maximum leaf width were constructed using Logistic equation. Based on simulation models, every day shape data of various varieties of winter wheat were calculated. With the help of OpenGL and NURBS surface modeling technology, winter wheat geometry model was built. Finally, combination of winter wheat growth model and shape model was realized, and growth process visualization of winter wheat after turning green stage was realized. [ Result ] Under the different varieties and different nitrogen levels, R2 of leaf length regression equation was between 0.772-0.983, F was between 10.153-340.191, and Sig was less than 0.05, R2 of maximum leaf width regression equation was between 0.853-0.999, F was between 17.371-4 359.236, and Sig was less than 0.05, the results showed that the model fitting degree and significance were better. After data validation, absolute error of leaf length model was between 0-3.88 cm, root mean squared error （RMSE） was between 0.24-1.95 cm, absolute error of maximum leaf width model was between 0-0.28 cm, and

关 键 词：冬小麦 有效积温 生长模型 形态模型 三维可视化 

分 类 号：S512.11[农业科学—作物学]