Markerska lokalizacija i registracija temeljena na medicinskom stereovizijskom sustavu

undergraduate thesis

• Title (english): Marker based localization and registration with a medical stereovision system
• Author: Bruno Gugo
• Mentor: Marko Švaco (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2024-09-24, Croatia
• Scientific / art field, discipline and subdiscipline: BIOMEDICINE AND HEALTHCARE; Public Health and Health Care; Public Health

Abstract

Procesi lokalizacije i registracije predstavljaju temelj današnjih računalno navođenih operacija.
Zadatak ovog završnog rada bio je uz pomoć medicinskog stereovizijskog sustava koji koristi krute markere najprije napraviti proces registracije prelaženjem ručno vođenog alata preko površine objekta, prilikom čega je generiran oblak točaka vezan uz koordinatni sustav od interesa, a zatim obaviti proces krute registracije, to jest pronalaženja matrice transformacije između koordinatnog sustava generiranog oblaka točaka te oblaka točaka objekta u prostoru slike (3D model). U početnim poglavljima objašnjen je princip rada stereovizijskog sustava te je napravljen pregled dostupnih metoda i algoritama za preklapanje dva oblaka točaka kao što su PCA za grubu registraciju te ICP za preciznu registraciju, ali i poboljšana verzija ICP metode Go-ICP algoritam. Na kraju je napravljena integracija svega te je provedena procjena točnosti sustava.

Abstract (english)

Processes of localization and registration represent the foundation of today's image-guided surgeries - IGS. The objective of this paper was to use a medical stereovision system that utilizes rigid markers to first perform the registration process by manually guiding a stylus over the surface of an object, during which a point cloud associated to the coordinate system of interest was generated. This was followed by a rigid registration process, i.e., finding the transformation matrix between the coordinate system of the generated point cloud and the point cloud of the object in the image space (3D model). The initial chapters explain the operating principle of the stereovision system and provide an overview of available methods and algorithms for overlapping two point clouds, such as PCA (Principal Component Analysis) for rough registration and ICP (Iterative Closest Point) for precise registration, along with the improved version of the ICP method, the Go-ICP algorithm (Globally Optimal – ICP). Finally, an integration of all the components was performed, and the accuracy of the system was evaluated.

• Keywords: računalno navođene operacije; registracija; lokalizacija; oblak točaka; PCA; ICP
• Keywords (english): image-guided surgeries (IGS); registration; localization; point cloud; PCA; ICP
• Language: croatian
• URN:NBN: urn:nbn:hr:235:882919
• Study programme: Title: Mechatronics and Robotics; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka mehatronike i robotike (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka mehatronike i robotike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-09-20 14:12:31