By Topic

Extracting Building Unit Number Information from High Resolution SAR Images with Regression Model

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Caixia Su ; Sch. of Math. & Comput. Sci., Guizhou Normal Univ., Guiyang, China ; Yongfeng Cao ; Jianjuan Liang

High resolution Synthetic Aperture Radar (SAR) sensor, which delivers images with metric or sub-metric resolution, makes it possible to extract detailed urban information. An effective method for extracting building unit number information from high resolution SAR images is proposed. In this method, a combination of intensity threshold and morphological operations are firstly used to detect buildings in SAR imagery. Then a regression function that describes the obvious correlativity existed between the features of the detected bright patch and the real number of buildings in the bright patch are modeled to predict the building number information in any region. The experiment on a TerraSAR_X image covering part of Wuhan city of China with spatial resolution 1.25×1.25m per pixel shows that the proposed method can get much more accurate building unit number information than using the number of the detected bright patches as the number of building units directly.

Published in:

Computational and Information Sciences (ICCIS), 2012 Fourth International Conference on

Date of Conference:

17-19 Aug. 2012