中国农业科学 ›› 2015, Vol. 48 ›› Issue (17): 3415-3428.doi: 10.3864/j.issn.0578-1752.2015.17.009

• 耕作栽培·生理生化·农业信息技术 • 上一篇    下一篇

数字植物研究进展:植物形态结构三维数字化

赵春江,陆声链,郭新宇,杜建军,温维亮,苗腾   

  1. 北京农业信息技术研究中心/国家农业信息化工程技术研究中心/农业部农业信息技术重点开放实验室/数字植物北京市重点实验室,北京 100097
  • 收稿日期:2015-05-04 出版日期:2015-09-01 发布日期:2015-09-01
  • 通讯作者: 赵春江,Tel:010-51503411;E-mail:zhaocj@nercita.org.cn
  • 作者简介:赵春江,Tel:010-51503411;E-mail:zhaocj@nercita.org.cn
  • 基金资助:
    国家自然科学基金(31171454)、国家“863”高技术研究发展计划资助(2013AA102405)、北京市科技新星计划资助(XX2013022)

Advances in Research of Digital Plant: 3D Digitization of Plant Morphological Structure

ZHAO Chun-jiang, LU Sheng-lian, GUO Xin-yu, DU Jian-jun, WEN Wei-liang, MIAO Teng   

  1. Beijing Research Center for Information Technology in Agriculture/National Engineering Research Center for Information Technology in Agriculture/Key Laboratory for Information Technology in Agriculture, Ministry of Agriculture/Beijing Key Laboratory of Digital Plant, Beijing 100097
  • Received:2015-05-04 Online:2015-09-01 Published:2015-09-01

摘要: 数字植物围绕农林植物生命、生产和生态系统的多维信息高效感知和认知的理论、技术和方法,通过多学科交叉合作,研究农林植物-环境3D数字化、高通量信息获取、情景感知、信息融合、结构和功能模拟、数字化设计和精准管理决策等数字农业的关键性、基础性以及共性理论和技术问题。植物形态结构的三维数字化是数字植物研究的重要组成部分,近年来很多学者从植物组织、器官、植株和群体等不同尺度,或者从植物根系和地上部等不同视角,围绕植物形态结构的参数测量、几何结构解析、三维模型构建、结构与功能建模,以及三维植物模型真实感展现等需要,开展了更深入的研究。在组织尺度方面,随着MRI、CT、显微成像等技术产品的不断成熟,使得利用这些先进测量仪器获取和测量植物组织内部结构数据成为可能,并被越来越多的研究者采用,成为进行植物内部形态结构测量和分析的有效手段。在植物根系的三维数字化方面,由于植物的根普遍生长在土壤里,观察和测量十分困难。虽然近年来XCT、MRI等穿透射线成像技术已越来越多地用于根系的形态结构探测,但这类技术往往仅能获取范围较小的根系局部数据,且价格昂贵。因此植物根系形态结构的准确、无损(原位)、快速测量仍然是一个挑战。在群体尺度方面,基于实测数据的三维重建逐渐成为植物群体三维重建的主要途径,研究者正试图从激光三维扫描仪获取的植物群体三维点云中提取群体的形态参数并实现群体的三维重构。而在三维植物模型的真实感绘制方面,如何准确地测量各种植物器官的光学特性并建立相应的数学模型是当前的研究重点,虽然已有不少研究者提出了相应的解决方案,但这些方法在便捷性和普适性方面仍然难以令人满意,有待更多深入研究。笔者最后结合相关领域的技术进展对数字植物的进一步研究进行了展望。

关键词: 数字植物, 植物形态结构建模, 三维重建, 可视化

Abstract: Digital plant focuses on research of the theories, technologies, and methods of efficient perception and cognition to the multi-dimensional information in plant life, agricultural production, and ecological system. The critical, basic, common theories, and technical problems in digital agriculture, including the 3D digitization of plants and its environment, high-throughput information acquisition, context awareness, information fusion, structural-functional simulation, digital design, and accurate management decision, can be studied in depth through the interdisciplinary cooperation. The study of 3D digitization to plant structure is one of the most important topics in the area of digital plant. Recently many in-depth studies on the 3D digitization of plant structure had been reported by considering the real requirements from parameter measurement of plant structure, analysis of morphological structure, 3D reconstruction, structural-functional modeling and realistic rendering of 3D plant models. These studies were done not only from tissue, organ, plant to colony scale, but also from aboveground to underground. In organ scale, with the advance of MRI, CT, microscopic imaging techniques and the corresponding products, using these advanced instruments to measure internal structure data of plant has become possible and popular in more and more researchers. And this also provides an effective method for measuring and analyzing the microcosmic morphological structure of plant. To the 3D digitization of root system, the observation and measurement of plant root are very difficult for root always grows deep in soil. Although XCT, MRI and X-ray imaging techniques have been more and more widely used in recent years for detecting the morphological structure of root, these techniques can often only be used to measure a small part of a whole root. So the accurate, nondestructive and fast measurement of root structure is still a challenge. At colony scale, real measured data-based 3D reconstruction has become a main method for the three-dimensional reconstruction of plant colony. Currently, some researchers are trying to extract the morphological parameters of plant colony directly from 3D point cloud then realize the 3D reconstruction. While at the research of realistic rendering of 3D plant model, how to accurately measure the optical properties of various plant organs and to establish a corresponding mathematical model is the focus in this area. However, although some solutions had been reported, these proposed methods are still not satisfactory when considering the convenience and general applicability, and more in-depth studies are expected to be done. Lastly, the further study of digital plant was also discussed based on the analysis of the related technical progress.

Key words: digital plant, modeling of plant morphology and structure, 3D reconstruction, visualization