Marker Detection for Surgical Navigation

Description: This Master Thesis describes the foundations, development, design and Implementation of the necessary algorithms to estimate the position and orientation of a marker of a surgical instrument tracked by a camera used in medical operations. This project presents a solution based on QR Codes attached to instruments and detected by a high de?nition camera in a surgical environment. QR markers contain information about the identi?cation of the instruments and their sizes, which is necessary to calculate the position of this planar marker. This thesis starts detailing the theory related to the QR Codes (encoding, reading, etc) and the State or Art of the marker navigation. Then the process of recognition of the markers in an image is described. The desired information of this process is obtaining the Region of Interest (ROI) and the encoded information for each QR Code. After that, the POSE is calculated using two di?erent methods, the ?rst one based on the point correspondences of the marker and the second one based on an iterative template matching comparing the homography transformation. Also is included a brief description about the software implementation. The algorithms are evaluated, giving as ?nal results the RMS rotation and translation errors for the computed POSE. From the evaluation, in optimum conditions the system achieves an accuracy of 5.80 [mm] in translation and 1.24 o in rotation.

Get full text: http://repositorio.educacionsuperior.gob.ec/handle/28000/2005