Modeling and optimization of 3D object shape recognition processes using neural networks
Students Name: Mykhailiuk Volodymyr Vasylovych
Qualification Level: magister
Speciality: Information Technology Design
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2023-2024 н.р.
Language of Defence: англійська
Abstract: Mykhailiuk V.V., Zdobytskyi A.Y. (supervisor). Modeling and optimization of 3D object shape recognition processes using neural networks. – Lviv Polytechnic National University, Lviv, 2023. Extended abstract. In the modern world, the role of unmanned vehicles is increasing every year. However, depending on the conditions of use, it is not always possible to directly control them by a person. Thus, the autonomy of unmanned systems is one of the challenges of today. One of the most important functions to ensure the autonomy of an unmanned system is its ability to recognize the shape of 3D objects without human intervention. 3D object shape recognition includes obtaining the geometry of the object (scanning the object), optimization of the obtained model and, in fact, recognition. Study object – modeling and optimization of 3D objects shape recognition processes using neural networks. Scope of research – methods of 3D objects shape recognition and optimization. Goal of research – optimization of the methodology for 3D objects shape recognition using neural networks. In the course of the Master’s thesis were studied the existing methods of 3D objects shape recognition. Based on the study, were formed the work requirements. Means for modeling and optimization of 3D objects shape recognition processes have been determined. A comparison of means for obtaining object geometry, creating a neural network for further optimization and 3D objects shape recognition was performed. Keywords: 3D objects shape recognition, optimization, 3D modeling, 3D scanning, neural network.