【作者】 陈郁淦； 周学成； 罗锡文； 曹霞； 乐凯；

【Author】 Yugan Chen1,Xuecheng Zhou1,Xiwen Luo1,Xia Cao1,Kai Le1(1.Laboratory of Key Technology on Agricultural Machine and Equipment and Equipment of South China Agricultural University,Ministry of Education,Guangzhou 510640,China)

【机构】 华南农业大学工程学院南方农业机械与装备关键技术教育部重点实验室；

【摘要】 本文采用目前国内外医学图像处理与分析领域运用最广泛的两个的算法平台:分割与配准工具箱ITK(Insight Segmentation and Registration Toolkit)和可视化工具箱VTK(Visualization Toolkit),对植物根系原位CT序列图像进行分割及三维可视化研究。首先,采用基于置信连接的区域生长算法对根系原位CT序列图像中的目标区域进行分割,并对分割后的图像进行数学形态学修整,以求更加完整地从断层图像中提取根系区域。然后,利用的移动立方体(Marching Cubes)算法对分割好的根系CT序列图像进行三维重建。最后,采用面向对象的编程语言C++结合ITK和VTK工具箱编程实现上述相关算法。实验结果表明本文提出的方法能够有效地实现根系CT序列图像的分割与三维重建,从而实现了对生长在介质环境中根系进行原位、快速、无损地三维可视化观测。更多还原

【Abstract】 With ITK(Insight Segmentation and Registration Toolkit)and VTK(Visualization Toolkit)widely used in the field of medical image processing and analysis platform in home and abroad,the segmentation and 3D visualization of plant root CT image sequences in situ were researched.First of all,the object region in CT image sequences of plant root was segmented by region growing algorithm based on confidence connection,and was corrected by mathematical morphology,so that it was extracted more complete from CT image sequences.And then,3D reconstruction of the CT image sequences segmented was realized by marching cubes algorithm.Finally,combined programming language C++ with ITK & VTK toolboxes,the algorithms above were implemented in Microsoft Visual Studio 6.0 programming environment.The experiment results show that the method put forward in this paper can realize the segmentation and 3D visualization of CT image sequences with high-speed,no damage and in situ.更多还原