计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (32): 208-210.

• 图形、图像、模式识别 • 上一篇    下一篇

采用特征跟踪的裂缝缺陷三维重建

贾 超,段亮亮,李 亚,张小敏   

  1. 燕山大学 信息科学与工程学院,河北 秦皇岛 066004
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-11-11 发布日期:2011-11-11

3D reconstruction of crack using feature tracking

JIA Chao,DUAN Liangliang,LI Ya,ZHANG Xiaomin   

  1. College of Information Science and Engineering,Yanshan University,Qinhuangdao,Hebei 066004,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-11-11 Published:2011-11-11

摘要: 通过对连续切片图像中裂缝区域跟踪实现裂缝缺陷的三维重建。分为以下三个步骤:区域预测、区域变形和表面绘制。预测是根据当前切片图像区域对下一个切片图像的区域进行预估。由于预测的结果和实际区域不一致采用区域变形技术得到实际的区域。对得到的区域进行表面绘制完成三维重建。在连续的切片图像中对提取出的区域不需对应关系的计算且在交互过程中利于特征的选取,因此计算成本低。实验结果表明通过上述方法可以对单个裂缝区域进行快速的跟踪和重建。

关键词: 特征跟踪, 卡尔曼滤波器, 区域变形, 三维重建

Abstract: Three-dimensional reconstruction of crack defects is achieved by tracking the crack region in the serial sections of the image.The process is divided into the following three steps:the regional prediction,regional deformation and surface rendering.Prediction is intended to estimate the region of following slice images based on the current image.It can get the actual region through region morphing due to the difference between the predicted region and the actual region.The region is applied surface rendering to complete the three-dimensional reconstruction.The need for explicit correspondence analysis between crack regions extracted independently in successive CT image slices is thus averted.The method is low cost,and thus facilitates interactive feature selection.The result shows that the crack region can be tracked and reconstructed quickly.

Key words: feature-based tracking, Kalman filter, region morphing, three-dimentional reconstruction