A new method of building footprints detection using airborne laser scanning data and multispectral image

Yiping Luo; Ting Jiang; Shengli Gao; Xin Wang

doi:10.1117/12.866946

22 October 2010 A new method of building footprints detection using airborne laser scanning data and multispectral image

Yiping Luo, Ting Jiang, Shengli Gao, Xin Wang

Author Affiliations +

Proceedings Volume 7658, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Detector, Imager, Display, and Energy Conversion Technology; 76581S (2010) https://doi.org/10.1117/12.866946
Event: 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies, 2010, Dalian, China

Abstract

It presents a new approach for detecting building footprints in a combination of registered aerial image with multispectral bands and airborne laser scanning data synchronously obtained by Leica-Geosystems ALS40 and Applanix DACS-301 on the same platform. A two-step method for building detection was presented consisting of selecting 'building' candidate points and then classifying candidate points. A digital surface model(DSM) derived from last pulse laser scanning data was first filtered and the laser points were classified into classes 'ground' and 'building or tree' based on mathematic morphological filter. Then, 'ground' points were resample into digital elevation model(DEM), and a Normalized DSM(nDSM) was generated from DEM and DSM. The candidate points were selected from 'building or tree' points by height value and area threshold in nDSM. The candidate points were further classified into building points and tree points by using the support vector machines(SVM) classification method. Two classification tests were carried out using features only from laser scanning data and associated features from two input data sources. The features included height, height finite difference, RGB bands value, and so on. The RGB value of points was acquired by matching laser scanning data and image using collinear equation. The features of training points were presented as input data for SVM classification method, and cross validation was used to select best classification parameters. The determinant function could be constructed by the classification parameters and the class of candidate points was determined by determinant function. The result showed that associated features from two input data sources were superior to features only from laser scanning data. The accuracy of more than 90% was achieved for buildings in first kind of features.

Citation Download Citation

Yiping Luo, Ting Jiang, Shengli Gao, and Xin Wang "A new method of building footprints detection using airborne laser scanning data and multispectral image", Proc. SPIE 7658, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Detector, Imager, Display, and Energy Conversion Technology, 76581S (22 October 2010); https://doi.org/10.1117/12.866946

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available