×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Proceedings, XXth congress


RECONSTRUCTION OF BUILDINGS FROM A SINGLE UAV IMAGE
WANG Jizhou*, Lin Zongjian®, LI Chengming®
? Institute of GIS, Chinese Academy of Surveying and Mapping, No.16, Road Beitaiping, District Haidian, Beijing, P.R.China,
100039, E-mail: wang116(2163.net, cmli()casm.ac.cn
? Chinese Academy of Surveying and Mapping, No.16, Road Beitaiping, District Haidian, Beijing, P.R.China, 100039,
E-mail: casm(d)casm.ac.cn
KEY WORDS: Photogrammetry; Acquisition; Measurement; GIS; Modeling
ABSTRACT:
Today we are up against the puzzle that how to acquire the 3D information of city efficiently and quickly. Many researchers
distributed all over the world put forward different methods towards their own applications, but there is still lack of a approach
which can meet most demands from various users. In this paper, we design and implement a 3D reconstruction method of city
buildings from single UAV(Unmanned Aerial Vehicle) image. As a new platform of Remote Sensing, UAV has many good
characteristics such as low cost, simple manipulation, high resolution, flexibility and etc, so we choose it as the tool for city image
acquisition. We can extract both geometry and texture information from UAV images. In a single UAV image, at first we get the
inner elements (xo, vo) and f by checking up the camera, secondly we divide the parallel lines in buildings into three groups which
parallel X, Y and Z coordinate axis respectively and compute the three joint points based on related photogrammetry theory, thirdly
calculate the three angle elements 4, aandkto implement the image relative orientation, fourthly compute the scale between reality
and image using the real coordinate information drawn from 2D GIS databases, then on the basis of all the above parameters we can
measure the real height of buildings and reconstruct their 3D models. In the paper, we have also build some accurate models to test
the feasibility and practicability of this method.
1. INTRODUCTION b. Extracting the layer information form the city map, and using
assumed layer height and virtual texture to reconstruct buildings.
Digital Earth and Digital City have become the research focuses The method is easy and needs no additional information. The
in geographic information field. High-resolution remote sensing main shortcoming is too artificial and lack of facticity.
images are the most important sources for the city 3D
information acquisition, and there are increasing demands. The c. Overlapping the digital orthoimage on the digital elevation
general remote sensing techniques by satellite and aircraft can model can easily found the three-dimensional scene of large
get the geographic information of large area, but there are some area city. But this scene is superficial ,the user can do scanning
limitations in detail information acquisition of city. The satellite and analyzing on a single object.
is restricted by the running cycle, height, cloud and some other
factors. And the aircraft is also affected by many things such as d. Digital photogrammetry. From the aerial remote sensing
airspace limit and weather. So theses techniques can’t image, we can derive plenty of geometric, texture and topologic
completely meet the require of subtle 3D information information by the means of digital photogrammetry, and it is
acquisition in city. Unmanned Air Vehicle, and the acronym is one of the most important method to reconstruct the 3D models
UAV, is a new platform of Remote Sensing. There is no man in of buildings.
UAV, and it is navigated by the telecontrol device. UAV has
many good characteristics such as low cost, simple e. Integration of multiple means.
manipulation, high resolution, flexibility and etc. UAV can
implement many tasks that traditional aircraft can’t finish. Lots — [n the paper, we put forward a building reconstruction method
of practices prove that UAV is a ideal tool for high-resolution from a single UAV image. We use.the UAVRS-II to get the
photos collection of city. high resolution image of city, then analyze every image and
derive the height and texture of buildings with the aid of digital
The city is the main inhabitation of people, which is mostly map in existence, at last reconstruct the corresponding 3D
composed of buildings. And how to factually describe the models.
building is the core task of Digital City. Researchers distributed
all over the world put forward various means to reconstruct
buildings as the following. 2. IMAGE ACQUISITION USING UAVRS-II
a. Using the design drawing of building and the city map The UAVRS-II remote sensing system is developed by the
together. This means is only fit for few buildings reconstruction, Chinese Academy of Surveying and Mapping which has
because it needs drawing recognition and lots of data input integrated the RS and UAV technology. The system comprises
work. And the building drawing is always very complex, it of RS device, RS platform and tele-controlling system. RS
needs professional knowledge to understand. device is used to acquire the photo, such as the CCD camera
which we used in the paper experiments. The platform is the
100
Inte
cal
rei
int
fly
ou
In
ne:
ani
los
ael
ob
as
tyr
sm
im
lin