DocumentCode :
144235
Title :
Urban building extraction via visual graphical topic model
Author :
Yansheng Li ; Yihua Tan ; Jinwen Tian
Author_Institution :
Nat. Key Lab. of Sci. & Technol. on Multi-spectral Inf. Process., Huazhong Univ. of Sci. & Technol., Wuhan, China
fYear :
2014
fDate :
13-18 July 2014
Firstpage :
4746
Lastpage :
4749
Abstract :
This paper addresses the automatic building extraction problem from high-resolution remote sensing images. The buildings in remote sensing images generally represent different shapes (i.e., simple rectangular or complex hybrid shape), it is intractable to extract all the buildings with different shapes once. Therefore, we adopt a hierarchical extraction style with a visual graphical topic model embedded, which includes two stages: the first stage detects the simple rectangular buildings and the second stage extracts the complex hybrid buildings. More specifically, the first stage is mainly responsible for regular buildings detection, and unsupervised visual graphical topic model (i.e., replicated softmax restricted boltzmann machine) and supervised discriminative model learning, and the second stage is mainly in charge of complex buildings extraction using the learned semantic feature mapping and discriminative model. Experimental results show that the second stage can obviously improve the building detection rate with slightly increasing the false alarm rate.
Keywords :
buildings (structures); geophysical image processing; image representation; image resolution; remote sensing; unsupervised learning; automatic building extraction problem; hierarchical extraction style; high-resolution remote sensing imaging; image representation; regular building detection; replicated soft max restricted Boltzmann machine; semantic feature mapping; supervised discriminative model learning; unsupervised visual graphical topic model; urban building extraction; Buildings; Data mining; Feature extraction; Remote sensing; Shape; Vegetation mapping; Visualization; Replicated softmax restricted boltzmann machine; hierarchical building extraction; visual topic; visual word;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International
Conference_Location :
Quebec City, QC
Type :
conf
DOI :
10.1109/IGARSS.2014.6947554
Filename :
6947554
Link To Document :
بازگشت