The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008
3. CONCLUSIONS.
The method processed the statistics of building main direction,
and planned the segments in accordance with the main direction,
so we can able to control the shape of the top of buildings good.
Taking the straight line of detected edge as reference, it has a
relative strong anti-noise capability. This method is relatively
stable, high accurate. Extracting is more dependent on the effect
of edge detecting, if the edge is not clear, shadow and the
surrounding features and buildings are very alike, the detection
result is not satisfied or the outcome is wrong. In addition,
during the line extension because it is difficult to find a method
of parallel extent, which can only divide region into rectangle,
and there are no unique guidelines of merger, results in the
extraction incomplete to the multi-angle type.
REFERENCE
Zhang Yu, Zhang Zuxun, Zhang Jianqing, J., 2000 The fast
Semi-automation extraction of buildings depends on
Geometrical restriction and image segmentation.. Transaction of
Wu Han Survey Scientific University, Vol.25(3), pp. 239-242
Yang Yijun, Zhao Rongchun, WangWenbing,J.2003 The
Automatic detect of artificial buildings on image,pp.214-218
Tao Wenbing, LiuJian.,J, 2001A new automatic extraction of
urban buildings on navigate image, 26(7), pp.867-873
TianYan, Zhangjun, Tao Wenbing,J, 2002 The Extraction of
city geometrical Information. Institute of Image Recognition
and Artificial Intelligence of HuaZhong Scienfic University.
Technical Report. ,30(4),pp, 120-124
ChenGuang„J,2003 Building recognition and change detection
on remote sensing image. Theory of master ’ s degree of
NanKing Science University,35(5)pp. 200
ACKNOWLEDGEMENT
Our research project i s supported by the “ National S cientific
Fund Program (No. 40771159)”, the “ University d octor
disciplines S cientific Fund Program of Ministry of Education
(No. 20070 147008 )” , the “Open Research Fund Program of
the State Key Laboratory of Information Engineering in
Surveying, Mapping and Remote Sensing of Wuhan
University (No. WKL(07)0303 )” , and the “42nd Postdoctor
S cientific Fund Program (No. 20070420918 )” .
1008