Title :
Roof-top detection based on structural elements combination
Author :
Feng, Hao ; Jiang, Zhiguo ; Yin, Jihao
Author_Institution :
Image Process. Center, Beijing Univ. of Aeronaut. & Astronaut., Beijing, China
Abstract :
A novel roof-top extraction method for satellite images based on probabilistic topic model is presented. We model roof-top as the connected structural elements. The proposed method contains two major steps: 1) Detect structural elements, different from earlier structure detector, the proposed method automatically learn the types of elements from unlabeled samples; 2) Connect these elements to form roof-top boundary, where the relationships between elements are estimated by hierarchical topic model. This approach belongs to generative method where only a small number of roof-top samples are required. The experimental results demonstrate the effectiveness of the proposed approach.
Keywords :
image sampling; image sensors; object detection; probability; remote sensing; roofs; structural engineering computing; hierarchical topic estimation model; probabilistic topic model; remote sensing image; roof-top extraction method; roof-top sample detection; satellite imaging; structural element detection combination; Computational modeling; Dictionaries; Indexes; Probabilistic logic; Remote sensing; Visualization; LDA; remote sensing; roof-top detection; topic model;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2012 IEEE International
Conference_Location :
Munich
Print_ISBN :
978-1-4673-1160-1
Electronic_ISBN :
2153-6996
DOI :
10.1109/IGARSS.2012.6351059
