【作者】 夏丹；

【导师】 李德华；

【作者基本信息】 华中科技大学 ， 控制科学与工程， 2012， 博士

【摘要】 分析植物的形态结构对研究植物的生长规律具有重要的理论意义,对农业生产中选种、配种、套种和提高农业产量等有着重要的指导作用。随着计算机技术的快速发展,三维信息获取技术使获取植物形态的三维数字化模型成为可能,而这将为研究植物的生理形态、结构和功能提供非常有用的直观分析工具。本论文旨在研制一整套植物形态三维扫描系统,该系统能够建立植物形态的三维数字化模型,并在计算机中进行三维显示。其主要工作过程为,将植物固定在转台上,由计算机控制转台旋转,同时使用摄像机拍摄植物的全方位二维数字图像,采用基于图像的三维建模算法对采集的序列图像进行处理,得到植物的表面点的空间坐标信息,生成点云文件,将点云三角网格化,建立三维模型,显示三维立体模型,并将三维数字化结果生成标准格式文件以方便读取和显示。本论文设计了系统的总体方案,搭建了硬件平台,为图像采集和获取植物形态的三维数字化模型提供了必须的硬件条件；完成了软件系统,能够对硬件设备和扫描过程进行控制,实现了系统各个功能模块,包含有图像采集模块,定标模块、三维重建模块和三维显示模块。其中,本论文着重对三维扫描系统中的关键技术即三维重建算法开展了深入研究。真实植物形态的三维几何外形能够用可见外壳来表征,计算可见外壳的算法分为基于体索的三维重建算法和基于表面的三维重建算法两大类。基于体素的算法较简单稳定但重建结果不够准确,而基于表面的算法的重建结果较精确但鲁棒性不强。为了在利用了基于表而重建的这类方法的优点的同时,保持基于体素重建的这类方法的简单和稳健性,本文提出了一种基于可见边的三维重建快速算法。该方法首先提取植物形态的轮廓,计算每个轮廓顶点对应的可见射线,然后将三维空间中的相交问题映射到二维图像平而,计算可见边,最后由可见边抽取植物形态的三维数字化模型。该方法能够快速地重建出相对精确的三维数字化模型,实验证明了方法的重建速度较快,该方法适合于需要实时建立植物形态的三维数字化模型的应用场合。为了得到较精确的三维数字化模型,本文提出了一种基于线模型的三维重建准确算法。该方法从多个侧影轮廓重建出八叉树结构表示的包围盒,对每个边缘体索来建立线模型,计算出线模型上在轮廓上连续的点,由这些点构造出物体的三维数字化模型。该方法是基于体素的三维重建方法,但是它也采用了基于表面的三维重建方法的思想,通过线模型直接计算可见外壳的表面点。该方法综合了基于体素的三维重建方法和基于表面的三维重建方法的优点,避免了各自的缺点,能够重建出精确的三维数字化模型,并且较稳健,实验证明了该方法的重建结果精度较高,该方法适合于需要建立植物形态的精确三维数字化模型的应用场合。建立植物根茎的三维数字化模型在精细化农业中有着重要意义,为了快速地建立非常精准的植物根茎的三维数字化模型,本文提出了一种适用于重建植物根茎的三维重建快速算法。该方法属于基于体素的重建可见外壳的方法。该方法以植物的根茎的三维骨架上的点来构造线模型,这些基于植物根茎的三维骨架而构造的线模型表征了植物根茎所在的三维空间,并有效地将其离散化。根据图像中轮廓的连续性来切割每个线模型,以此来获取位于线模型上位于根茎的表而上的点。该方法能够兼顾速度和准确度,能在较短的时间内重建出精确的植物根茎的三三维数字化模型。实验证明了该方法速度快、精度高,重建结果逼近植物根茎的真实形态。更多还原

【Abstract】 The research of morphology of plants has important theoretical significance in the research of study of plant growth pattern. The research plays an important role in guiding agricultural production, seed selection, breeding, intercropping, and increasing agricultural production. With the rapid development of computer technology, three-dimensional information acquisition technology can reconstruct three-dimensional digital model for plant morphology. The three-dimensional information acquisition technology provides a very useful visual analysis tools for studying plant physiological form, structure and function.A 3D scanning system for the plant morphology based on the non-contact surveying is introduced. It acquires 3D geometric information of the plant morphology surface through software processing with the body shape captured by visual equipments. The system is a high-tech product of abroad application. It can acquire automatically the coordinate information of the object surface points in 3D space, build 3D models of plant morphology surface, and save the acquired data as corresponding standard interface files. Several key problems of the 3D scanning system for the plant are emphasized:overall technical scheme of the system, software analysis of the system, spatial information acquisition from the scanned plant morphology, and so on.A systematical overview for the current status, technique points and future development trend of the art of 3D information acquisition technology at home and abroad is introduced. Hardware system and software modules of the system are summarized and detailed analysis on collection, pre-processing and acquisition of the 3D spatial information are deeply discussed, too.The method of the system is based on the idea of Image Based Modeling and Visual Hull which could satisfy all the practical working requirements with a low cost. The methods computing visual hulls are mainly classified into two categories:surface-based approaches and volume-based approaches. Surface-based approaches are precise and lack robustness while volume-based approaches are robust but inaccurate and work slowly. The paper proposes a novel method to compute visual hull rapidly. To this aim, the method is based on viewing edges. First, the viewing lines are computed from all the contours of the object from different viewpoints. Viewing edges are computed by simplifying the 3D intersection onto 2D image plane. Finally, a water-tight surface is extracted from the surface points on the visual hull. Experiments with real data are conducted to test the rapidity of the proposed method.A novel method is proposed to compute exact visual hull by taking advantage of surface-based approaches while retaining the simplicity and robustness of volume-based approaches. To this aim, a line-based model is built for each cube on the edge of octree volume structure of object. Convex surface points of visual hull in line-based model are searched by the idea of surface-based approaches. Experiments with real data are conducted to test the efficiency and exactness of our method.Modeling the 3D shape of plant stems is very important for the study of plant growth in precision agriculture. To construct 3D model of real plant stems from images fast, a novel volumetric method based on line-based models is proposed in this paper. The line-based models are constructed on the coarse 3D skeleton of the plant stems. Then the line-based models are carved with respect to silhouette consistency. The surface points on the plant stems are calculated from line-based models. Finally, a mesh surface model can be extracted from the surface points. The proposed method can give precise results together with low time complexity and space complexity. Experiments based on both synthetic data and real data are presented to evaluate the preciseness and spcediness of the proposed method.更多还原