Tridemensional reconstruction of scenes with LiDAR and camera

• Autor(a) principal: Lourenço, Bernardo Miguel Martins
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/29112
• topic: 3D reconstruction; Laser scanner; Point cloud; Calibration; Normal estimation; Color fusion; Color registration

Tridimensional reconstruction is still a challenging area, that has multiple application in architecture and robotics. Several technologies are used today, like Stereoscopy or Structured Light, however, none is able to achieve precise geometric results, which are usually required. A technology, LiDAR, has evolved as the de facto technology for tridimensional reconstruction, being able to achieve unmatched results. Yet, this technology is unable to register the color of objects, so the usual solution is to use a camera for this. Therefore, in this work, a set of algorithms and techniques for tridimensional reconstruction with a LiDAR laser scanner and a camera was developed. Moreover, a 3D scanner was developed to register real-word scenes. In particular, an innovative calibration method was developed to calibrate the laser scanner, which performed above a similar calibration method. Finally, the reconstruction process was tested with real data. The geometric reconstruction was very accurate but the color reconstruction was not, especially because of the poor calibration of the camera.