Abstract:
3-D reconstruction of buildings in urban areas is one
of the highlighted issues in photogrammetry, remote
sensing, computer vision, and computer graphic. It can be
used in various fields such as urban planning,
construction, disaster management, navigation system,
and cyber city .This paper presents an approach for 3D
reconstruction model of buildings from urban area for 3D
visualization. The model is generated automatically from
a single image taken by UAV (Unmanned aerial vehicles).
Most methods used two or more images for 3D building
reconstruction. The previous methods take much time to
detect and match feature points in each image and
consume more memory space for multiple images. The
proposed system can extract 3D geometry information
from a single image without knowing the priori
parameters. The building features are extracted using
modified building feature extraction algorithm and then
reconstruct it 3D model using proposed 3D box method.
The proposed method can reduce processing time and
memory space by using a single image for 3D reconstruction.
