高级检索
肖伯祥, 秦洪. 玉米叶片物理建模与参数确定的运动学分析方法[J]. 计算机辅助设计与图形学学报, 2019, 31(5): 718-725. DOI: 10.3724/SP.J.1089.2019.17373
引用本文: 肖伯祥, 秦洪. 玉米叶片物理建模与参数确定的运动学分析方法[J]. 计算机辅助设计与图形学学报, 2019, 31(5): 718-725. DOI: 10.3724/SP.J.1089.2019.17373
Xiao Boxiang, Qin Hong. Physical Maize Leaf Modeling and Parameters Determination by Motion Analysis[J]. Journal of Computer-Aided Design & Computer Graphics, 2019, 31(5): 718-725. DOI: 10.3724/SP.J.1089.2019.17373
Citation: Xiao Boxiang, Qin Hong. Physical Maize Leaf Modeling and Parameters Determination by Motion Analysis[J]. Journal of Computer-Aided Design & Computer Graphics, 2019, 31(5): 718-725. DOI: 10.3724/SP.J.1089.2019.17373

玉米叶片物理建模与参数确定的运动学分析方法

Physical Maize Leaf Modeling and Parameters Determination by Motion Analysis

  • 摘要: 基于物理的方法是解决植物动态建模与仿真的有效途径,而模型参数的确定是基于物理方法所面临的重要问题.为实现基于生物物理属性特征的玉米叶片高精度、高真实感动态建模与仿真,提出一种基于运动学分析的玉米叶片物理建模与参数确定方法.首先利用运动捕捉技术获取玉米叶片的运动数据;然后根据运动捕捉特征点构建基于质点-弹簧系统的物理模型;最后基于运动学分析建立物理模型参数解析方程组,利用最小二乘法计算物理模型的参数.以此参数构建物理模型实现玉米叶片的动态模拟,并给出2种模型构造方式的玉米叶片虚拟仿真.结果表明,该方法以一种数据驱动的机制有效估算出对应的物理模型参数,生成的玉米叶片动态虚拟仿真过程具有与实际情况较为接近的真实感效果,为虚拟植物建模与应用提供了一种可行的物理建模及参数确定方法.

     

    Abstract: Physics-based method is an effective way to solve the problem of dynamic plant leaf modeling and simulation. However, an important task is parameters determination of physical model. Aiming to high-accuracy and high-realistic dynamic modeling and simulation based on biophysical attributes, this paper describes an approach to physics-based modeling and parameters determination for maize leaf. First, we acquired the motion data of maize leaf by motion capture. Second, we constructed a mass-spring-based physical model according to the motion capture markers. And then, we constructed equations for parameters calculation based on motion analysis and got solution by least square. Finally, by use of calculated parameters, we performed dynamic simulation of maize leaf model and presented comparison of two types of models. The experimental results show that the proposed method determines the parameters of physical model by a data-driven way and achieve approximate and realistic simulation results, as a feasible modeling and parameter determination method for virtual plant modeling application.

     

/

返回文章
返回