Description

In image based 3D modeling either from UAVs or from the ground, incomplete coverage of the object of interest is quite happen either because of the mis-planning or the occlusions and accessibility constraints. Accordingly, the produced 3D model might be incomplete and missing some parts or gaps. Therefore, to have a complete and adequate documentation the gaps should be identified and then recovered by auxiliary image data in a second round epoch or in a one-go campaign.

Objectives and Methodology

In this MSc project, the objective is to apply an automated and efficient detection of gaps in the 3D reconstructed object parts and replan the imaging network to have finally a fully coved object with the required level of details and accuracy. One option is to detect the gaps by digitally classifying the point clouds or the source images using AI, then additional UAV waypoints can be designed for a final capture and completeness.

Further reading