Bi-directional gradient labeling and registration for gray-scale image segmentation

Lei Ma, Xiao Ping Zhang, Jennie Si, Glen P. Abousleman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Watershed is one of the commonly used methods for image segmentation. In this paper, we introduce a new segmentation scheme based on bi-directional labeling and registration and prove that its segmentation performance is equivlent to that of conventional watershd but it is much more computational efficent. The bi-directional labeling and registration scheme, which will be referred to as BIDS, involves only linear scans of image pixels. It uses one dimensional operations instead of queues while traditional segmentation algorithms are two dimensional problems. BIDS also provides unique labels for each homogeneous regions. In addition to achieving the same segmentation results as conventional watershed, BIDS is four times less computationally complex than the conventional watersheds by immersion.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Image Processing
Pages365-368
Number of pages4
Volume1
StatePublished - 2003
EventProceedings: 2003 International Conference on Image Processing, ICIP-2003 - Barcelona, Spain
Duration: Sep 14 2003Sep 17 2003

Other

OtherProceedings: 2003 International Conference on Image Processing, ICIP-2003
CountrySpain
CityBarcelona
Period9/14/039/17/03

Fingerprint

Watersheds
Image segmentation
Labeling
Labels
Pixels

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Ma, L., Zhang, X. P., Si, J., & Abousleman, G. P. (2003). Bi-directional gradient labeling and registration for gray-scale image segmentation. In IEEE International Conference on Image Processing (Vol. 1, pp. 365-368)

Bi-directional gradient labeling and registration for gray-scale image segmentation. / Ma, Lei; Zhang, Xiao Ping; Si, Jennie; Abousleman, Glen P.

IEEE International Conference on Image Processing. Vol. 1 2003. p. 365-368.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ma, L, Zhang, XP, Si, J & Abousleman, GP 2003, Bi-directional gradient labeling and registration for gray-scale image segmentation. in IEEE International Conference on Image Processing. vol. 1, pp. 365-368, Proceedings: 2003 International Conference on Image Processing, ICIP-2003, Barcelona, Spain, 9/14/03.
Ma L, Zhang XP, Si J, Abousleman GP. Bi-directional gradient labeling and registration for gray-scale image segmentation. In IEEE International Conference on Image Processing. Vol. 1. 2003. p. 365-368
Ma, Lei ; Zhang, Xiao Ping ; Si, Jennie ; Abousleman, Glen P. / Bi-directional gradient labeling and registration for gray-scale image segmentation. IEEE International Conference on Image Processing. Vol. 1 2003. pp. 365-368
@inproceedings{fb4e36751983409a9504f903c1c32456,
title = "Bi-directional gradient labeling and registration for gray-scale image segmentation",
abstract = "Watershed is one of the commonly used methods for image segmentation. In this paper, we introduce a new segmentation scheme based on bi-directional labeling and registration and prove that its segmentation performance is equivlent to that of conventional watershd but it is much more computational efficent. The bi-directional labeling and registration scheme, which will be referred to as BIDS, involves only linear scans of image pixels. It uses one dimensional operations instead of queues while traditional segmentation algorithms are two dimensional problems. BIDS also provides unique labels for each homogeneous regions. In addition to achieving the same segmentation results as conventional watershed, BIDS is four times less computationally complex than the conventional watersheds by immersion.",
author = "Lei Ma and Zhang, {Xiao Ping} and Jennie Si and Abousleman, {Glen P.}",
year = "2003",
language = "English (US)",
volume = "1",
pages = "365--368",
booktitle = "IEEE International Conference on Image Processing",

}

TY - GEN

T1 - Bi-directional gradient labeling and registration for gray-scale image segmentation

AU - Ma, Lei

AU - Zhang, Xiao Ping

AU - Si, Jennie

AU - Abousleman, Glen P.

PY - 2003

Y1 - 2003

N2 - Watershed is one of the commonly used methods for image segmentation. In this paper, we introduce a new segmentation scheme based on bi-directional labeling and registration and prove that its segmentation performance is equivlent to that of conventional watershd but it is much more computational efficent. The bi-directional labeling and registration scheme, which will be referred to as BIDS, involves only linear scans of image pixels. It uses one dimensional operations instead of queues while traditional segmentation algorithms are two dimensional problems. BIDS also provides unique labels for each homogeneous regions. In addition to achieving the same segmentation results as conventional watershed, BIDS is four times less computationally complex than the conventional watersheds by immersion.

AB - Watershed is one of the commonly used methods for image segmentation. In this paper, we introduce a new segmentation scheme based on bi-directional labeling and registration and prove that its segmentation performance is equivlent to that of conventional watershd but it is much more computational efficent. The bi-directional labeling and registration scheme, which will be referred to as BIDS, involves only linear scans of image pixels. It uses one dimensional operations instead of queues while traditional segmentation algorithms are two dimensional problems. BIDS also provides unique labels for each homogeneous regions. In addition to achieving the same segmentation results as conventional watershed, BIDS is four times less computationally complex than the conventional watersheds by immersion.

UR - http://www.scopus.com/inward/record.url?scp=0344666652&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0344666652&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0344666652

VL - 1

SP - 365

EP - 368

BT - IEEE International Conference on Image Processing

ER -