【24h】

Multimodal Image Registration and Visual Servoing

机译:多模式图像配准和视觉伺服

获取原文

摘要

This paper deals with multimodal imaging in the surgical robotics context. On the first hand, it addresses numerical registration of a preoperative image obtained by fluorescence with an intraoperative image grabbed by a conventional white-light endoscope. This registration involves displacement and rotation in the image plane as well as a scale factor. On the second hand, a method is developed to visually servo the endoscope to the preoperative imaging location. Both methods are original and dually based on the use of mutual information between a pair of fluorescence and white-light images and of a modified Nelder-Mead simplex algorithm. Numerical registration is validated on real images whereas visual servoing is validated experimentally in two set-ups: a planar microrobotic platform and a 6DOF parallel robot.
机译:本文涉及手术机器人背景中的多模式成像。在第一手上,它地址通过荧光通过传统的白光内窥镜捕获的术中图像获得的术前图像的数值登记。该注册涉及在图像平面中的位移和旋转以及比例因子。在二手中,开发了一种方法以在视觉上伺服内窥镜到术前成像位置。这两种方法都是原始的,并且基于在一对荧光和白光图像之间使用互信息以及改进的Nelder-MeD单纯x算法。在实际图像上验证了数值注册,而目视伺服以两种设置实验验证:平面微生物平台和6dof并联机器人。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号