Matching segmentation algorithms to ERS-1 SAR applications

Ronald G. Caves; Shaun Quegan

doi:10.1117/12.197534

21 December 1994 Matching segmentation algorithms to ERS-1 SAR applications

Ronald G. Caves, Shaun Quegan

Proceedings Volume 2316, SAR Data Processing for Remote Sensing; (1994) https://doi.org/10.1117/12.197534
Event: Satellite Remote Sensing, 1994, Rome, Italy

Abstract

We compare the performance of two segmentation algorithms, previously proven on high resolution airborne synthetic aperture radar (SAR) data, on lower resolution spaceborne SAR data. The algorithms are: RWSEG - iterative multiscale edge detection/segment growing; and ANNEAL - maximum a posteriori radar cross-section reconstruction via simulated annealing. To test the utility and robustness of the algorithms they are applied to ERS-1 PRI images of an agricultural area in the UK and salt playa in Tunisia. These scenes are chosen for the differences between what we would define as useful segmentations of them. A number of tests are applied to segmentation output to measure the homogeneity of segments and the complexity of segment boundaries. The segmentations produced by ANNEAL are generally more detailed than those produced by RWSEG, but take much longer to produce. We also investigate how RWSEG can be used to detect structural change in multitemperal sequences of images. It is found that it is not possible to clearly identify structural similiarites and differences when images are segmented separately and segment boundaries are then overlaid. Much better results occur when a multitemporal sequence is segmented as a single entity.

Citation Download Citation

Ronald G. Caves and Shaun Quegan "Matching segmentation algorithms to ERS-1 SAR applications", Proc. SPIE 2316, SAR Data Processing for Remote Sensing, (21 December 1994); https://doi.org/10.1117/12.197534

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