Watershed segmentation with automatic altitude selection and region merging based on the markov random field model

Wen Feng Kuo, Yung Nien Sun

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Watershed transformation has proven to be an important tool in image analysis. However, the resulting image of watershed transformation is inevitably over-segmented due to the presence of noise or local irregularities in the input image. In this paper, the use of contour altitude at the immersion stage is proposed. Block gradient information computed from the input gradient image is defined and used to obtain a critical altitude of watershed flooding. This altitude is then refined based on entropy estimated from the intermediate segmentation result. Thereafter, an optimal altitude and its corresponding segmentation result can be obtained. Although this process can favorably reduce the number of regions, the quality of segmentation still requires further improvement. Hence, a Markov Random Field (MRF) model defined on a region adjacency graph (RAG) is adopted. Because the MRF model can merge neighboring regions that are similar in local statistic properties, it thus alleviates the over-segmentation problem and improves the quality of image segmentation. In the experimental studies, the proposed method has been tested using several benchmark images. It achieves improved appearance and energy indices in comparison with the results obtained by conventional methods.

Original languageEnglish
Pages (from-to)153-171
Number of pages19
JournalInternational Journal of Pattern Recognition and Artificial Intelligence
Volume24
Issue number1
DOIs
Publication statusPublished - 2010 Feb 1

Fingerprint

Watersheds
Merging
Image segmentation
Image analysis
Entropy
Statistics

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{bd914e751fb34233b57d762679aaee45,
title = "Watershed segmentation with automatic altitude selection and region merging based on the markov random field model",
abstract = "Watershed transformation has proven to be an important tool in image analysis. However, the resulting image of watershed transformation is inevitably over-segmented due to the presence of noise or local irregularities in the input image. In this paper, the use of contour altitude at the immersion stage is proposed. Block gradient information computed from the input gradient image is defined and used to obtain a critical altitude of watershed flooding. This altitude is then refined based on entropy estimated from the intermediate segmentation result. Thereafter, an optimal altitude and its corresponding segmentation result can be obtained. Although this process can favorably reduce the number of regions, the quality of segmentation still requires further improvement. Hence, a Markov Random Field (MRF) model defined on a region adjacency graph (RAG) is adopted. Because the MRF model can merge neighboring regions that are similar in local statistic properties, it thus alleviates the over-segmentation problem and improves the quality of image segmentation. In the experimental studies, the proposed method has been tested using several benchmark images. It achieves improved appearance and energy indices in comparison with the results obtained by conventional methods.",
author = "Kuo, {Wen Feng} and Sun, {Yung Nien}",
year = "2010",
month = "2",
day = "1",
doi = "10.1142/S021800141000783X",
language = "English",
volume = "24",
pages = "153--171",
journal = "International Journal of Pattern Recognition and Artificial Intelligence",
issn = "0218-0014",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "1",

}

TY - JOUR

T1 - Watershed segmentation with automatic altitude selection and region merging based on the markov random field model

AU - Kuo, Wen Feng

AU - Sun, Yung Nien

PY - 2010/2/1

Y1 - 2010/2/1

N2 - Watershed transformation has proven to be an important tool in image analysis. However, the resulting image of watershed transformation is inevitably over-segmented due to the presence of noise or local irregularities in the input image. In this paper, the use of contour altitude at the immersion stage is proposed. Block gradient information computed from the input gradient image is defined and used to obtain a critical altitude of watershed flooding. This altitude is then refined based on entropy estimated from the intermediate segmentation result. Thereafter, an optimal altitude and its corresponding segmentation result can be obtained. Although this process can favorably reduce the number of regions, the quality of segmentation still requires further improvement. Hence, a Markov Random Field (MRF) model defined on a region adjacency graph (RAG) is adopted. Because the MRF model can merge neighboring regions that are similar in local statistic properties, it thus alleviates the over-segmentation problem and improves the quality of image segmentation. In the experimental studies, the proposed method has been tested using several benchmark images. It achieves improved appearance and energy indices in comparison with the results obtained by conventional methods.

AB - Watershed transformation has proven to be an important tool in image analysis. However, the resulting image of watershed transformation is inevitably over-segmented due to the presence of noise or local irregularities in the input image. In this paper, the use of contour altitude at the immersion stage is proposed. Block gradient information computed from the input gradient image is defined and used to obtain a critical altitude of watershed flooding. This altitude is then refined based on entropy estimated from the intermediate segmentation result. Thereafter, an optimal altitude and its corresponding segmentation result can be obtained. Although this process can favorably reduce the number of regions, the quality of segmentation still requires further improvement. Hence, a Markov Random Field (MRF) model defined on a region adjacency graph (RAG) is adopted. Because the MRF model can merge neighboring regions that are similar in local statistic properties, it thus alleviates the over-segmentation problem and improves the quality of image segmentation. In the experimental studies, the proposed method has been tested using several benchmark images. It achieves improved appearance and energy indices in comparison with the results obtained by conventional methods.

UR - http://www.scopus.com/inward/record.url?scp=77951613825&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951613825&partnerID=8YFLogxK

U2 - 10.1142/S021800141000783X

DO - 10.1142/S021800141000783X

M3 - Article

AN - SCOPUS:77951613825

VL - 24

SP - 153

EP - 171

JO - International Journal of Pattern Recognition and Artificial Intelligence

JF - International Journal of Pattern Recognition and Artificial Intelligence

SN - 0218-0014

IS - 1

ER -