Image segmentation based on kernel PCA and shape prior

Xiaoping Wan; Djamal Boukerroui; Jean-Pierre Cocquerez

doi:10.1117/12.896324

8 July 2011 Image segmentation based on kernel PCA and shape prior

Xiaoping Wan, Djamal Boukerroui, Jean-Pierre Cocquerez

Proceedings Volume 8009, Third International Conference on Digital Image Processing (ICDIP 2011); 800937 (2011) https://doi.org/10.1117/12.896324
Event: 3rd International Conference on Digital Image Processing, 2011, Chengdu, China

Abstract

The introduction of shape priori in the segmentation model ameliorates effectively the poor segmentation result due to the using of the image information alone to segment the image including noise, occlusion, or missing parts. But the presentation of shape via Principal Component Analysis (PCA) brings on the limitation of the similarity between the objet and the prior shape. In this paper, we proposed using Kernel PCA (KPCA) to capture the shape information - the variability. KPCA can present better shape prior knowledge. The model based on KPCA allows segmenting the object with nonlinear transformation or a quite difference with the priori shape. Moreover, since the shape model is incorporated into the deformable model, our segmentation model includes the image term and the shape term to balance the influence of the global image information and the shape prior knowledge in proceed of segmentation. Our model and the model based on PCA both are applied to synthetic images and CT medical images. The comparative results show that KPCA can more accurately identify the object with large deformation or from the noised seriously background.

Citation Download Citation

Xiaoping Wan, Djamal Boukerroui, and Jean-Pierre Cocquerez "Image segmentation based on kernel PCA and shape prior", Proc. SPIE 8009, Third International Conference on Digital Image Processing (ICDIP 2011), 800937 (8 July 2011); https://doi.org/10.1117/12.896324

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