A novel SAR fusion image segmentation method based on Markov Random Field

Author

Xu, Huaping ; Wang, Wei ; Liu, Xianghua

Author_Institution

Sch. of Electron. & Inf. Eng., Beijing Univ. of Aeronaut. & Astronaut., Beijing, China

Volume

3

fYear

2010

fDate

16-18 Oct. 2010

Firstpage

1297

Lastpage

1300

Abstract

Markov Random Field (MRF) method is a popular technology in SAR image segmentation nowadays. It considers the statistical characteristics of SAR image and achieves optimal image segmentation result. In this paper, a novel SAR fusion image segmentation method based on MRF model is proposed. Firstly, the mechanism of MRF segmentation on single SAR image is studied. Secondly, the Maximum a Posterior (MAP) formula for SAR fusion image segmentation is deduced by supposing the two SAR images for fusion are statistically independent. Then the energy function of SAR fusion image segmentation is presented and the processing steps are given. At the end, computer simulation indicates that the performance of this new approach is much better than that of single SAR image segmentation based on MRF.

Keywords

Markov processes; image segmentation; sensor fusion; synthetic aperture radar; Markov random field; energy function; maximum a posterior formula; synthetic aperture radar fusion image segmentation; Computational modeling; Computer simulation; Image resolution; Image segmentation; Markov random fields; Object detection; Pixel; Markov Random Field; SAR; data fusion; image segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing (CISP), 2010 3rd International Congress on

Conference_Location

Yantai

Print_ISBN

978-1-4244-6513-2

Type

conf

DOI

10.1109/CISP.2010.5647694

Filename

5647694