Bayesian segmentation for damage image using MRF prior

G. Li; F. G. Yuan; R. Haftka; N. H. Kim

doi:10.1117/12.816507

30 March 2009 Bayesian segmentation for damage image using MRF prior

G. Li, F. G. Yuan, R. Haftka, N. H. Kim

Proceedings Volume 7292, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2009; 72920J (2009) https://doi.org/10.1117/12.816507
Event: SPIE Smart Structures and Materials + Nondestructive Evaluation and Health Monitoring, 2009, San Diego, California, United States

Abstract

Image segmentation for quantifying damage based on Bayesian updating scheme is proposed for diagnosis and prognosis in structural health monitoring. This scheme enables taking into account the prior information of the state of the structures, such as spatial constraints and image smoothness. Bayes' law is employed to update the segmentation with the spatial constraint described as Markov Random Field and the current observed image acting as a likelihood function. Segmentation results demonstrate that the proposed algorithm holds promise of searching a crack area in the SHM image and focusing on the real damage area by eliminating the pseudo-shadow area. Thus more precise crack estimation can be obtained than the conventional K-means segmentation by shrinking the fuzzy tails which often exist on both sides of the crack tips.

Citation Download Citation

G. Li, F. G. Yuan, R. Haftka, and N. H. Kim "Bayesian segmentation for damage image using MRF prior", Proc. SPIE 7292, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2009, 72920J (30 March 2009); https://doi.org/10.1117/12.816507

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