Visual Marker Detection and Decoding in AR Systems: A Comparative Study
ISMAR '02 Proceedings of the 1st International Symposium on Mixed and Augmented Reality
Merging Visible and Invisible: Two Camera-Augmented Mobile C-Arm (CAMC) Applications
IWAR '99 Proceedings of the 2nd IEEE and ACM International Workshop on Augmented Reality
Multiple View Geometry in Computer Vision
Multiple View Geometry in Computer Vision
Parallax-Free Long Bone X-ray Image Stitching
MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part I
Inverse C-arm Positioning for Interventional Procedures Using Real-Time Body Part Detection
MICCAI '09 Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention: Part I
Hi-index | 0.00 |
X-ray images are widely used during surgery for long bone fracture fixation. Mobile C-arms provide X-ray images which are used to determine the quality of trauma reduction, i.e. the extremity length and mechanical axis of long bones. Standard X-ray images have a narrow field of view and can not visualize the entire long bone on a single image. In this paper, we propose a novel method to generate panoramic X-ray images in real time by using the previously introduced Camera Augmented Mobile C-arm [1]. This advanced mobile C-arm system acquires registered X-ray and optical images by construction, which facilitates the generation of panoramic X-ray images based on first stitching the optical images and then embedding the X-ray images. We additionally introduce a method to reduce the parallax effect that leads to the blurring and measurement error on panoramic X-ray images. Visual marker tracking is employed to automatically stitch the sequence of video images and to rectify images. Our proposed method is suitable for intra-operative usage generating panoramic X-ray images, which enable metric measurements, with less radiation and without requirement of fronto-parallel setup and overlapping X-ray images. The results show that the panoramic X-ray images generated by our method are accurate enough (errors less than 1%) for metric measurements and suitable for many clinical applications in trauma reduction.