The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-2/W11
https://doi.org/10.5194/isprs-archives-XLII-2-W11-533-2019
https://doi.org/10.5194/isprs-archives-XLII-2-W11-533-2019
04 May 2019
 | 04 May 2019

MAGO: A NEW APPROACH FOR ORTHOPHOTOS PRODUCTION BASED ON ADAPTIVE MESH RECONSTRUCTION

S. Gagliolo, B. Federici, I. Ferrando, and D. Sguerso

Keywords: Orthophoto, Point cloud, Mesh, Images, Algorithm, Image classification, Statistical analysis

Abstract. Orthophotos are one of the most common and typical products of a photogrammetric post-processing and, since the diffusion of specific software, their generation and usage have become even more widespread. In spite of it, some issues remain on the accuracy of orthophoto reconstruction, which is often downgraded by the introduction of meshes and Digital Surface Models to be used as surfaces representing the object. The use of a more accurate and reliable input, such as a point cloud, makes these approximations avoidable. For this reason, a new approach, termed MAGO (Adaptive Mesh for Orthophoto Reconstruction), is here delineated and proposed. The input data of the procedure are the user-defined orthophoto plane, the image and its internal and external orientation parameters, and a point cloud representing the object. Each pixel of the image is projected on the orthophoto plane at its original resolution via an iterative process, which builds an adaptive mesh, defined by means of the three best fitting points, where the collinearity rays and the point cloud intersect. After an overview on the method and its innovative features, an example on a test case is reported, together with a comparison between MAGO’s and another photogrammetric software results.