Integration of Heterogeneous Three-Dimensional Point Cloud Processing Algorithms

In recent years, three-dimensional imaging techniques have become an increasingly active research area. One of the most common outputs from a three-dimensional imaging system is a point cloud, a collection of points in three-dimensional space. Many techniques exist for generating point clouds, including image-based approaches and LIght Detection And Ranging (Lidar). Point clouds have become increasingly easy to generate and software libraries have emerged that allow researchers to work with point clouds in a more natural way. However, there is not one library that gives a researcher all of the tools that they need to develop novel point cloud algorithms. In this thesis, a new point cloud processing framework will be presented that allows for seamless integration of existing point cloud libraries and allows researchers to develop point cloud algorithms in a high-level setting. As a demonstration of the framework’s capabilities, a plane-fitting algorithm case study will be implemented.