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.
Chong WU
City University of Hong Kong
Le ZHANG
Tongji University
Houwang ZHANG
China University of Geosciences
Hong YAN
City University of Hong Kong
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
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 -