Share Email Print
cover

Proceedings Paper

Research on remote sensing image segmentation based on ant colony algorithm: take the land cover classification of middle Qinling Mountains for example
Author(s): Xin Mei; Qian Wang; Quanfang Wang; Wenfang Lin
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Remote sensing image based on the complexity of the background features, has a wealth of spatial information, how to extract huge amounts of data in the region of interest is a serious problem. Image segmentation refers to certain provisions in accordance with the characteristics of the image into different regions, and it is the key of remote sensing image recognition and information extraction. Reasonably fast image segmentation algorithm is the base of image processing; traditional segmentation methods have a lot of the limitations. Traditional threshold segmentation method in essence is an ergodic process, the low efficiency impacts on its application. The ant colony algorithm is a populationbased evolutionary algorithm heuristic biomimetic, since proposed, it has been successfully applied to the TSP, job-shop scheduling problem, network routing problem, vehicle routing problem, as well as other cluster analysis. Ant colony optimization algorithm is a fast heuristic optimization algorithm, easily integrates with other methods, and it is robust. Improved ant colony algorithm can greatly enhance the speed of image segmentation, while reducing the noise on the image. The research background of this paper is land cover classification experiments according to the SPOT images of Qinling area. The image segmentation based on ant colony algorithm is carried out and compared with traditional methods. Experimental results show that improved the ant colony algorithm can quickly and accurately segment target, and it is an effective method of image segmentation, it also has laid a good foundation of image classification for the follow-up work.

Paper Details

Date Published: 30 October 2009
PDF: 8 pages
Proc. SPIE 7494, MIPPR 2009: Multispectral Image Acquisition and Processing, 74942F (30 October 2009); doi: 10.1117/12.833156
Show Author Affiliations
Xin Mei, Hubei Univ. (China)
Qian Wang, Hubei Univ. (China)
Quanfang Wang, Hubei Univ. (China)
Wenfang Lin, Hubei Univ. (China)


Published in SPIE Proceedings Vol. 7494:
MIPPR 2009: Multispectral Image Acquisition and Processing
Faxiong Zhang; Faxiong Zhang, Editor(s)

© SPIE. Terms of Use
Back to Top