A new method of extracting shape features from IKONOS imagery based on Fourier Descriptor: an application to object-oriented classification

Wei Wan; Xuezhi Feng; Pengfeng Xiao; Limin Zhao

doi:10.1117/12.872947

3 November 2010 A new method of extracting shape features from IKONOS imagery based on Fourier Descriptor: an application to object-oriented classification

Wei Wan, Xuezhi Feng, Pengfeng Xiao, Limin Zhao

Author Affiliations +

Proceedings Volume 7840, Sixth International Symposium on Digital Earth: Models, Algorithms, and Virtual Reality; 78401I (2010) https://doi.org/10.1117/12.872947
Event: The Sixth International Symposium on Digital Earth, 2009, Beijing, China

Abstract

Shape is an important visual feature of very high-resolution satellite data. Fourier Descriptor (FDs) was introduced in this paper as a new method to extract and represent objects' shape features of IKONOS imagery and a 5-dimensional (5- D) feature-vector was proposed as a shape parameter. A classification model was established based on K-means clustering algorithm, the 5-D feature-vector was taken as discrimination variable together with the mean gray values. The results showed that when involving the shape feature-vector into the classification model, the overall classification accuracy was 82.4% with 84.6% producer accuracy of roads. So it was confirmed a feasible way to represent shape features of remotely sensed imagery based on FDs.

Citation Download Citation

Wei Wan, Xuezhi Feng, Pengfeng Xiao, and Limin Zhao "A new method of extracting shape features from IKONOS imagery based on Fourier Descriptor: an application to object-oriented classification", Proc. SPIE 7840, Sixth International Symposium on Digital Earth: Models, Algorithms, and Virtual Reality, 78401I (3 November 2010); https://doi.org/10.1117/12.872947

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