University of Twente Student Theses


Perception algorithms for localization of a pneumatic endoscope

Rusu, L.I. (2022) Perception algorithms for localization of a pneumatic endoscope.

[img] PDF
Abstract:The project aims to help increase the level of safety and efficiency of colonoscopies by tracking the tip of the colonoscope and localizing it in terms of the intestinal anatomy. While there are visual, electromagnetic, or other position detection solutions currently in use, they mostly focus on colons with typical anatomy, whereas this project aims to take into consideration difficult scenarios, such as deformation or poor quality of intestinal features. Using as the primary method a monocular Simultaneous Localization and Mapping (SLAM) algorithm, this research focuses on localizing the distal end of the colonoscope visually, in terms of the surrounding features. Based on a literature comparison of multiple SLAM algorithms, ORB-SLAM (Oriented FAST and Rotated BRIEF SLAM) was considered the best alternative. The validation of the results is done with an Aurora electromagnetic sensor. The report will start with a general context and a literature review that serve as the starting point for choosing the methods. Furthermore, the approach is described. After a theoretical overview of the employed methods (ORB-SLAM3, blob detection, camera calibration), the report will contain a description of the software and hardware, as well as the experimental setup. Finally, the results are presented, along with a discussion on limitations and future work, and conclusions.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Electrical Engineering MSc (60353)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page