Height-Gradient-Based Method for Occlusion Detection in True Orthophoto Generation
MetadataShow full item record
High-quality orthophotos are essential elements in cartographic databases that are required for engineering projects. Urban areas usually have many tall buildings, causing occlusions in aerial images. These occlusions can be severe depending on the height of the buildings, geometry of the imaging system, and image acquisition viewpoint. A product where no occlusion areas appear is called true orthophoto, and occlusion detection is a key step in its generation. This process requires a set of aerial images and surface representation, together with metadata and acquisition system information. This letter addresses a new approach for occlusion detection. The central idea that allows the development of the proposed method is based on the analysis of the surface height gradient at certain sampled directions, guiding the identification of occluded regions in the aerial images. A novel metric is introduced based on the height gradient computed along an available digital surface model, which can be obtained by different techniques. The refinement of the occlusion area is made by applying one morphological operator, aiming to fill some gaps in the detected occlusion areas. In order to validate the proposed approach, simulated and real data were used. The quality assessment was based on both visual and numerical analyses. For this methodology, the numerical analysis showed high completeness and correctness values, indicating that the proposed approach works properly.