Advanced search
Start date
Betweenand
Related content
(Reference retrieved automatically from SciELO through information on FAPESP grant and its corresponding number as mentioned in the publication by the authors.)

Orientation and orthorectification of nadiral terrestrial images

Full text
Author(s):
Adilson Berveglieri [1] ; Marcus Vinícius Antunes de Moraes [2] ; Antonio Maria Garcia Tommaselli [3]
Total Authors: 3
Affiliation:
[1] Universidade Estadual Paulista. Faculdade de Ciências e Tecnologia,. Departamento de Cartografia - Brasil
[2] Universidade Estadual Paulista. Faculdade de Ciências e Tecnologia,. Departamento de Cartografia - Brasil
[3] Universidade Estadual Paulista. Faculdade de Ciências e Tecnologia,. Departamento de Cartografia - Brasil
Total Affiliations: 3
Document type: Journal article
Source: Bol. Ciênc. Geod.; v. 21, n. 1, p. 149-168, 2015-03-00.
Abstract

Ground control information is essential for photogrammetric applications. No matter the type of control, a manual interactive step is still required to locate image coordinates of Ground Control Points (GCPs). This approach proposes an automatic technique to orientate terrestrial images of GCPs and to generate ortho-images. These ortho-images can be used as control scenes in matching algorithms to orientate aerial and orbital images. The method uses a control target attached to the Global Navigation Satellite System (GNSS) receiver while a GCP is surveyed. GCP images are collected in nadir view and at different heights. A multi-scale model is generated and oriented using tie points and the target as control. The tie point coordinates are automatically determined by intersection of rays. A local Digital Terrain Model (DTM) is produced to orthorectify the area with centre at the GCP. Experiments were performed using small areas with elevation variations. The estimated 3D coordinates were compared to corresponding samples collected by topographic measurements, and the results showed a root mean square errors less than 2 cm in Z coordinate for the reconstructed areas, which is compatible with the target applications. (AU)

FAPESP's process: 10/16954-5 - Development of image matching techniques for multi-scale and multi-optical sensors georeferenced data
Grantee:Adilson Berveglieri
Support Opportunities: Scholarships in Brazil - Doctorate