Deep ArUco: AI/ML-based Real-Time Marker Pose Tracking
dc.contributor.advisor | Kristian Muri Knausgård | |
dc.contributor.advisor | Andreas Klausen | |
dc.contributor.advisor | Sondre Sanden Tørdal | |
dc.contributor.author | Hebnes, Silje Wetrhus | |
dc.date.accessioned | 2022-09-26T16:24:38Z | |
dc.date.available | 2022-09-26T16:24:38Z | |
dc.date.issued | 2022 | |
dc.identifier | no.uia:inspera:109927222:70437059 | |
dc.identifier.uri | https://hdl.handle.net/11250/3021479 | |
dc.description.abstract | Machine learning is commonly used in varoius types of machine vision. Convolutional neural network (CNN) are models that can be trained in different lighting, colors, changes and motion blur. This study generates data containing images with ArUco markers to be detected in different real-world scenarios. Environments created to detect the markers in this study is different lighting, motion blur, rain drops, different contrasts and fog. The ArUco markers will be detected by an existing detection algorithm using machine learning and artificial intelligence. | |
dc.description.abstract | ||
dc.language | ||
dc.publisher | University of Agder | |
dc.title | Deep ArUco: AI/ML-based Real-Time Marker Pose Tracking | |
dc.type | Master thesis |
Tilhørende fil(er)
Denne innførselen finnes i følgende samling(er)
-
Master's theses in Mechatronics [105]
MAS500