Image segmentation by an encoder-segmented neural network

Ning Li; Youfu Li

doi:10.1117/12.317659

31 July 1998 Image segmentation by an encoder-segmented neural network

Ning Li, Youfu Li

Proceedings Volume 3384, Photonic Processing Technology and Applications II; (1998) https://doi.org/10.1117/12.317659
Event: Aerospace/Defense Sensing and Controls, 1998, Orlando, FL, United States

Abstract

Most of region-based image segmentation approaches suffered from the problem of different thresholds selecting for different images. In this paper, an new adaptive image segmentation approach based on an encoder-segmented neural network (ESNN) is presented. The novel ESNN combines the advantages of self-organizing feature map (SOFM) and fuzzy c- means clustering (FCM) algorithm. Feature encoder implemented by SOFM for vector quantization using the competitive learning where the feature vectors can be encoded as the definite sequence by which the most of the available feature vectors can be extracted for the final segmentation using encoded feature-based fuzzy c-means (EFFCM) algorithm. Since the contribution of feature encoder, ESNN can reduce the complexion of computation when processing a large number of multi-spectral images. ESNN have been applied for brain MRI segmentation. Comparing with FCM algorithm, experimental results have shown ESNN method for segmentation makes better performance on computation and adaptability.

Citation Download Citation

Ning Li and Youfu Li "Image segmentation by an encoder-segmented neural network", Proc. SPIE 3384, Photonic Processing Technology and Applications II, (31 July 1998); https://doi.org/10.1117/12.317659

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