项目名称: 基于多轴飞行器航拍数据的植物生长分析与建模
项目编号: No.61502060
项目类型: 青年科学基金项目
立项/批准年度: 2016
项目学科: 计算机科学学科
项目作者: 刘骥
作者单位: 重庆大学
项目金额: 20万元
中文摘要: 如何简化植物功能结构模型的构建过程是植物生长建模的重要问题。通过分析植物生长的视频数据,进而重建植物功能结构模型,被证明是解决该问题的有效手段。但如何在自然环境下获取和分析描述植物生长的视频数据,仍然是尚未解决的问题。本项目围绕该问题开展研究,通过控制多轴飞行器实现对户外植物生长过程的定期、定点拍摄,获取植物生长的视频数据。基于这些数据,利用物体识别和运动到结构方法,重建植物形态的3D点云,形成描述植物生长过程的4D点云。经过对4D点云数据的分析,实现植物器官的标记、植物生理参数的提取、生长事件的识别、生长规则的导出,最终利用L系统自动化建立植物的功能结构模型,实现植物生长的建模。
中文关键词: 植物建模;多轴飞行器;三维重建;计算机视觉;功能结构植物模型
英文摘要: How to simplify the process of building Functional Structural Plant Model is a key issue of plat growth modeling. Although reconstructing Functional Structural Plant Model from analyzing videos of plant growth has been proved in practice to be a very effective means, how to capture and analyze such kind of videos in the wild are still open questions. To address these problems, this project plans to use multirotor to make videos periodically at fixed position for a plant as it grow. Then 4D cloud points can be reconstructed from a series of 3D cloud points, which are acquired by analyzing videos of a plant based on object recognition and structure from motion technologies. This project aims to design a framework which can labeling plant organs, extracting physiological parameters, detecting growth event and exporting growth rules from the 4D cloud points. Finally with these informations, the Functional Structural Plant Model of a plant can be built automatically using L-system, and the plant growth modeling can be realized.
英文关键词: Plant Modeling;Multirotor;3D Reconstruction;Computer Vision;Functional Structural Plant Model