Rekonstrukcija objekata metodom registracije oblaka točaka

master's thesis

• Title (english): Point cloud registration for object reconstruction
• Author: Davor Gregurić
• Mentor: Matko Orsag (mentor)
• Committee member: Matko Orsag (predsjednik povjerenstva)
• Committee member: Zdenko Kovačić (član povjerenstva)
• Committee member: Goran Vasiljević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-09-23, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Rad se bavi rekonstrukcijom objekata u obliku 3D modela u virtualnom okruženju, korištenjem registracije oblaka točaka kao osnove. Sustav se sastoji od PlantScan uređaja za 3D skeniranje i rekonstrukciju objekata, primarno biljaka, pomoću RGB-D kamere, i pripadne programske podrške. Rješavanje problema registracije oblaka točaka podijeljeno je na probleme transformacije položaja kamere, pohrane oblaka i registracije oblaka u ROS okruženju. Problem transformacije riješen je dodavanjem funkcija za računanje i odašiljanje transformacije u skriptu za pomicanje motora iz sustava PlantScan. Problem pohrane oblaka riješen je korištenjem skripte za pohranu slika i oblaka objekta u rosbag zapis. Na kraju, problem registracije oblaka i rekonstrukcije u 3D model riješen je korištenjem dobivenih transformacija i oblaka točaka kako bi se, pomoću metode iterativne najbliže točke, oblaci spojili u 3D model. U sklopu rada prikazana je teorijska podloga navedenih problema i njihovih rješenja.

Abstract (english)

The thesis deals with object reconstruction in the form of 3D models in virtual environment, using point cloud registration as basis. The system consists of the PlantScan device for 3D object scanning and reconstruction, with plants as primary subject, via RGB-D camera and the accompanying software. Resolving the problem of point cloud registration is further divided into the problems of camera position transform, point cloud storage and cloud registration in ROS environment. The transform problem is solved by incorporating the transform calculation and broadcast functions into the motor movement script from the PlantScan system. The storage problem is solved using the script for object image and cloud storage in the rosbag format. Finally, the cloud registration and 3D model reconstruction problem is solved using the obtained transforms and point clouds to merge them into a 3D model with the aid of the iterative closest point method. The thesis includes theoretical background for all of the aforementioned problems and their solutions.

• Keywords: oblak točaka; rekonstrukcija objekata; 3d model; iterativna najbliža točka; ICP
• Keywords (english): pointcloud; object reconstruction; 3d model; iterative closest point; ICP
• Language: croatian
• URN:NBN: urn:nbn:hr:168:210771
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 09:07:33