A TSVM based semi-supervised approach to SAR Image Segmentation

Ji Jun; Shao Fengjing; Sun Rencheng; Zhang Neng; Liu Guanfeng

doi:10.1109/ETTandGRS.2008.13

A TSVM based semi-supervised approach to SAR Image Segmentation

Ji Jun^*, Shao Fengjing, Sun Rencheng, Zhang Neng, Liu Guanfeng

^*Corresponding author for this work

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

10 Citations (Scopus)

Abstract

Image segmentation is a fundamental issue in image processing. Segmentation of synthetic aperture radar (SAR) images is extremely difficult on account of intrinsic multiplicative speckle noises. Due to the ambiguities of SAR images, labeled instances are difficult and time-consuming to obtain while unlabeled data are abundant. In this paper, a new semi-supervised approach based on transductive support vector machine (TSVM) is proposed to segment SAR images, it is robust to noises and is effective when dealing with low numbers of high-dimensional samples, moreover, it could efficiently make use of unlabeled data to reduce human labor and improve precision. Segmentation results are also compared to SVM and TSVM trained by using different samples and parameters. Experimental results demonstrate that the proposed method is very promising.

Original language	English
Title of host publication	2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008
Pages	495-498
Number of pages	4
Volume	1
DOIs	https://doi.org/10.1109/ETTandGRS.2008.13
Publication status	Published - 2009
Event	2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008 - Shanghai, China Duration: 21 Dec 2008 → 22 Dec 2008

Other

Other	2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008
Country/Territory	China
City	Shanghai
Period	21/12/08 → 22/12/08

Keywords

Image segmentation
SAR
Semisupervised
TSVM

Access to Document

10.1109/ETTandGRS.2008.13

Cite this

@inproceedings{1b8a4fd51e634d168fc8f52951aad8be,

title = "A TSVM based semi-supervised approach to SAR Image Segmentation",

abstract = "Image segmentation is a fundamental issue in image processing. Segmentation of synthetic aperture radar (SAR) images is extremely difficult on account of intrinsic multiplicative speckle noises. Due to the ambiguities of SAR images, labeled instances are difficult and time-consuming to obtain while unlabeled data are abundant. In this paper, a new semi-supervised approach based on transductive support vector machine (TSVM) is proposed to segment SAR images, it is robust to noises and is effective when dealing with low numbers of high-dimensional samples, moreover, it could efficiently make use of unlabeled data to reduce human labor and improve precision. Segmentation results are also compared to SVM and TSVM trained by using different samples and parameters. Experimental results demonstrate that the proposed method is very promising.",

keywords = "Image segmentation, SAR, Semisupervised, TSVM",

author = "Ji Jun and Shao Fengjing and Sun Rencheng and Zhang Neng and Liu Guanfeng",

year = "2009",

doi = "10.1109/ETTandGRS.2008.13",

language = "English",

isbn = "9780769535630",

volume = "1",

pages = "495--498",

booktitle = "2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008",

note = "2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008 ; Conference date: 21-12-2008 Through 22-12-2008",

}

Jun, J, Fengjing, S, Rencheng, S, Neng, Z & Guanfeng, L 2009, A TSVM based semi-supervised approach to SAR Image Segmentation. in 2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008. vol. 1, 5070204, pp. 495-498, 2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008, Shanghai, China, 21/12/08. https://doi.org/10.1109/ETTandGRS.2008.13

A TSVM based semi-supervised approach to SAR Image Segmentation. / Jun, Ji; Fengjing, Shao; Rencheng, Sun et al.
2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008. Vol. 1 2009. p. 495-498 5070204.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - A TSVM based semi-supervised approach to SAR Image Segmentation

AU - Jun, Ji

AU - Fengjing, Shao

AU - Rencheng, Sun

AU - Neng, Zhang

AU - Guanfeng, Liu

PY - 2009

Y1 - 2009

N2 - Image segmentation is a fundamental issue in image processing. Segmentation of synthetic aperture radar (SAR) images is extremely difficult on account of intrinsic multiplicative speckle noises. Due to the ambiguities of SAR images, labeled instances are difficult and time-consuming to obtain while unlabeled data are abundant. In this paper, a new semi-supervised approach based on transductive support vector machine (TSVM) is proposed to segment SAR images, it is robust to noises and is effective when dealing with low numbers of high-dimensional samples, moreover, it could efficiently make use of unlabeled data to reduce human labor and improve precision. Segmentation results are also compared to SVM and TSVM trained by using different samples and parameters. Experimental results demonstrate that the proposed method is very promising.

AB - Image segmentation is a fundamental issue in image processing. Segmentation of synthetic aperture radar (SAR) images is extremely difficult on account of intrinsic multiplicative speckle noises. Due to the ambiguities of SAR images, labeled instances are difficult and time-consuming to obtain while unlabeled data are abundant. In this paper, a new semi-supervised approach based on transductive support vector machine (TSVM) is proposed to segment SAR images, it is robust to noises and is effective when dealing with low numbers of high-dimensional samples, moreover, it could efficiently make use of unlabeled data to reduce human labor and improve precision. Segmentation results are also compared to SVM and TSVM trained by using different samples and parameters. Experimental results demonstrate that the proposed method is very promising.

KW - Image segmentation

KW - SAR

KW - Semisupervised

KW - TSVM

UR - https://www.scopus.com/pages/publications/70350635378

U2 - 10.1109/ETTandGRS.2008.13

DO - 10.1109/ETTandGRS.2008.13

M3 - Conference proceeding contribution

SN - 9780769535630

VL - 1

SP - 495

EP - 498

BT - 2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008

T2 - 2008 International Workshop on Education Technology and Training and 2008 International Workshop on Geoscience and Remote Sensing, ETT and GRS 2008

Y2 - 21 December 2008 through 22 December 2008

ER -

A TSVM based semi-supervised approach to SAR Image Segmentation

Abstract

Other

Keywords

Access to Document

Other files and links

Fingerprint

Cite this