Key points are not available for this paper at this time.
Increasing number of skyscrapers along with the large number of tall bridges in the urban setting throughout the world also increases the demand of robust and autonomously controlled structural health monitoring (SHM) in order to enhance the reliability of such structures and their surroundings. In this paper, an unmanned aerial vehicle (UAV)-based autonomous SHM system has been investigated, in which the images of the structural site captured by the UAV were stitched together to form the complete view of the structure. The image-stitching task has been done by using a well-known speeded up robust features (SURF)-based feature detection algorithm. The large number of features resulting from SURF are first reduced with random sample consensus algorithm and then the respective transformations are applied to align the images for final stitching. The comparison between the current and previous view of the structural site provides the structural differences. The proposed approach is tested on a sample structure in a lab with different possible realistic types of defects that are induced in the structure, and the performance of the proposed methodology is compared with the existing approaches. It has been shown that the proposed system can perform image stitching even if the UAV suffers angular displacement due to wind thrusts or calibration issues. The proposed approach has also been applied on a concrete structure, and the displacement detected on the column of the structure's backyard verified the feasibility for real-world SHM.
Akbar et al. (Tue,) studied this question.