Superpixel Based Hierarchical Segmentation for Color Image

Chong WU; Le ZHANG; Houwang ZHANG; Hong YAN

doi:10.1587/transinf.2020EDL8025

Superpixel Based Hierarchical Segmentation for Color Image

Chong WU, Le ZHANG, Houwang ZHANG, Hong YAN

Full Text Views

0

Cite this

Summary :

In this letter, we propose a hierarchical segmentation (HS) method for color images, which can not only maintain the segmentation accuracy, but also ensure a good speed. In our method, HS adopts the fuzzy simple linear iterative clustering (Fuzzy SLIC) to obtain an over-segmentation result. Then, HS uses the fast fuzzy C-means clustering (FFCM) to produce the rough segmentation result based on superpixels. Finally, HS takes the non-iterative K-means clustering using priority queue (KPQ) to refine the segmentation result. In the validation experiments, we tested our method and compared it with state-of-the-art image segmentation methods on the Berkeley (BSD500) benchmark under different types of noise. The experiment results show that our method outperforms state-of-the-art techniques in terms of accuracy, speed and robustness.

Publication: IEICE TRANSACTIONS on Information Vol.E103-D No.10 pp.2246-2249

Publication Date: 2020/10/01

Publicized: 2020/07/03

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2020EDL8025

Type of Manuscript: LETTER

Category: Image Processing and Video Processing

Authors

Chong WU
  City University of Hong Kong
Le ZHANG
  Tongji University
Houwang ZHANG
  China University of Geosciences
Hong YAN
  City University of Hong Kong

Keyword

image segmentation, robustness, superpixel

Cite this

Copy

Chong WU, Le ZHANG, Houwang ZHANG, Hong YAN, "Superpixel Based Hierarchical Segmentation for Color Image" in IEICE TRANSACTIONS on Information, vol. E103-D, no. 10, pp. 2246-2249, October 2020, doi: 10.1587/transinf.2020EDL8025.
Abstract: In this letter, we propose a hierarchical segmentation (HS) method for color images, which can not only maintain the segmentation accuracy, but also ensure a good speed. In our method, HS adopts the fuzzy simple linear iterative clustering (Fuzzy SLIC) to obtain an over-segmentation result. Then, HS uses the fast fuzzy C-means clustering (FFCM) to produce the rough segmentation result based on superpixels. Finally, HS takes the non-iterative K-means clustering using priority queue (KPQ) to refine the segmentation result. In the validation experiments, we tested our method and compared it with state-of-the-art image segmentation methods on the Berkeley (BSD500) benchmark under different types of noise. The experiment results show that our method outperforms state-of-the-art techniques in terms of accuracy, speed and robustness.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDL8025/_p

Copy

@ARTICLE{e103-d_10_2246,
author={Chong WU, Le ZHANG, Houwang ZHANG, Hong YAN, },
journal={IEICE TRANSACTIONS on Information},
title={Superpixel Based Hierarchical Segmentation for Color Image},
year={2020},
volume={E103-D},
number={10},
pages={2246-2249},
abstract={In this letter, we propose a hierarchical segmentation (HS) method for color images, which can not only maintain the segmentation accuracy, but also ensure a good speed. In our method, HS adopts the fuzzy simple linear iterative clustering (Fuzzy SLIC) to obtain an over-segmentation result. Then, HS uses the fast fuzzy C-means clustering (FFCM) to produce the rough segmentation result based on superpixels. Finally, HS takes the non-iterative K-means clustering using priority queue (KPQ) to refine the segmentation result. In the validation experiments, we tested our method and compared it with state-of-the-art image segmentation methods on the Berkeley (BSD500) benchmark under different types of noise. The experiment results show that our method outperforms state-of-the-art techniques in terms of accuracy, speed and robustness.},
keywords={},
doi={10.1587/transinf.2020EDL8025},
ISSN={1745-1361},
month={October},}

Copy

TY - JOUR
TI - Superpixel Based Hierarchical Segmentation for Color Image
T2 - IEICE TRANSACTIONS on Information
SP - 2246
EP - 2249
AU - Chong WU
AU - Le ZHANG
AU - Houwang ZHANG
AU - Hong YAN
PY - 2020
DO - 10.1587/transinf.2020EDL8025
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2020
AB - In this letter, we propose a hierarchical segmentation (HS) method for color images, which can not only maintain the segmentation accuracy, but also ensure a good speed. In our method, HS adopts the fuzzy simple linear iterative clustering (Fuzzy SLIC) to obtain an over-segmentation result. Then, HS uses the fast fuzzy C-means clustering (FFCM) to produce the rough segmentation result based on superpixels. Finally, HS takes the non-iterative K-means clustering using priority queue (KPQ) to refine the segmentation result. In the validation experiments, we tested our method and compared it with state-of-the-art image segmentation methods on the Berkeley (BSD500) benchmark under different types of noise. The experiment results show that our method outperforms state-of-the-art techniques in terms of accuracy, speed and robustness.
ER -