GPS/IMU based SIFT-matching for aerial romote sensing images

Author

Ji, Shunping ; Ma, Dan

Author_Institution

Sch. of Remote Sensing & Inf. Eng., Wuhan Univ., Wuhan, China

fYear

2011

fDate

24-26 June 2011

Firstpage

1257

Lastpage

1260

Abstract

The paper presents a new automatic matching method for aerial imagery based on SIFT with image exterior orientation (EO) elements obtained by GPS/IMU system. Firstly, the image EO elements are used to establish the epipolar equations for constraining y-parallax and to predict the initial x-parallax with approximate elevation. Secondly, the SIFT extraction strategy based on pyramid and slicing method are presented for the sake of large capacity of aerial imagery. Thirdly, the GPS/IMU epipolar equations are used to constrain the searching range in matching step and then the least square matching method are used to improve the matching accuracy to sub-pixel level. At last, the gross error detection method base on varying weight is used to eliminate the miss-matching features.

Keywords

Global Positioning System; geophysical image processing; least squares approximations; remote sensing; GPS-IMU system; SIFT-matching; aerial remote sensing images; automatic matching method; epipolar equations; gross error detection method; image exterior orientation; least square matching method; matching accuracy; Computer vision; Equations; Forestry; Global Positioning System; Image matching; Information science; Remote sensing; GPS/IMU; SIFT; aerial imagery; epipolar equation; image matching;

fLanguage

English

Publisher

ieee

Conference_Titel

Remote Sensing, Environment and Transportation Engineering (RSETE), 2011 International Conference on

Conference_Location

Nanjing

Print_ISBN

978-1-4244-9172-8

Type

conf

DOI

10.1109/RSETE.2011.5964508

Filename

5964508