Premium
True orthophoto generation using line segment matches
Author(s) -
Wang Qiang,
Yan Lei,
Sun Yanbiao,
Cui Ximin,
Mortimer Hugh,
Li Yanyan
Publication year - 2018
Publication title -
the photogrammetric record
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.638
H-Index - 51
eISSN - 1477-9730
pISSN - 0031-868X
DOI - 10.1111/phor.12229
Subject(s) - orthophoto , sawtooth wave , line (geometry) , line segment , computer science , feature (linguistics) , point cloud , point (geometry) , matching (statistics) , constraint (computer aided design) , artificial intelligence , computer vision , triangulated irregular network , algorithm , mathematics , geometry , geography , remote sensing , linguistics , philosophy , statistics , digital elevation model
When generating a true orthophoto from aerial urban scenes, especially those containing man‐made features with large height differences, sawtooth effects in feature edges can occur in the rectified images. Aiming to eliminate such effects, this study proposes an advanced orthorectification method using line segment matches, allowing 3D building edges to be accurately reconstructed. The corresponding 2D line segments are first extracted and matched, enabling the reconstruction of 3D line segments by joining two planes and imposing a line end‐point constraint. The 3D line segments are then dissected into discrete 3D points to be incorporated into the 3D point cloud obtained by a dense matching algorithm. Finally, a more complete and accurate triangulated irregular network ( TIN ) model can be constructed to provide important basic data for true orthophoto production. Experimental results show that sawtooth effects can be eliminated, resulting in significantly improved quality in the true orthophotograph.