Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation

Siyuan Chen; Yijun Yan; Jinchang Ren; Byongjun Hwang; Stephen Marshall; Tariq Durrani

doi:10.1007/978-981-19-0390-8_126

Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation

Siyuan Chen, Yijun Yan, Jinchang Ren, Byongjun Hwang, Stephen Marshall, Tariq Durrani

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

Abstract

By grouping pixels with visual coherence, superpixel algorithms provide an alternative representation of regular pixel grid for precise and efficient image segmentation. In this paper, a multi-stage model is used for sea ice segmentation from the high-resolution optical imagery, including the pre-processing to enhance the image contrast and suppress the noise, superpixel generation and classification, and post-processing to refine the segmented results. Four superpixel algorithms are evaluated within the framework, where the high-resolution imagery of the Chukchi sea is used for validation. Quantitative evaluation in terms of the segmentation quality and floe size distribution, and visual comparison for several selected regions of interest are presented. Overall, the model with TS-SLIC yields the best results, with a segmentation accuracy of 98.19% on average and adhering to the ice edges well.

Original language	English
Title of host publication	Communications, Signal Processing, and Systems
Subtitle of host publication	Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems
Editors	Qilian Liang, Wei Wang, Xin Liu, Zhenyu Na, Baoju Zhang
Publisher	Springer Singapore
Pages	1004-1012
Number of pages	9
Volume	1
ISBN (Electronic)	9789811903908
ISBN (Print)	9789811903892, 9789811903922
DOIs	https://doi.org/10.1007/978-981-19-0390-8_126
Publication status	Published - 31 Mar 2022
Event	10th International Conference on Communications, Signal Processing, and Systems - Changbaishan, China Duration: 24 Jul 2021 → 25 Jul 2021 Conference number: 10

Publication series

Name	Lecture Notes in Electrical Engineering
Publisher	Springer
Volume	878 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	10th International Conference on Communications, Signal Processing, and Systems
Abbreviated title	CSPS 2021
Country/Territory	China
City	Changbaishan
Period	24/07/21 → 25/07/21

Access to Document

10.1007/978-981-19-0390-8_126Licence: Unspecified

Cite this

Chen, S., Yan, Y., Ren, J., Hwang, B., Marshall, S., & Durrani, T. (2022). Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation. In Q. Liang, W. Wang, X. Liu, Z. Na, & B. Zhang (Eds.), Communications, Signal Processing, and Systems : Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems (Vol. 1, pp. 1004-1012). (Lecture Notes in Electrical Engineering; Vol. 878 LNEE). Springer Singapore. https://doi.org/10.1007/978-981-19-0390-8_126

Chen, Siyuan ; Yan, Yijun ; Ren, Jinchang et al. / Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images : Analysis and Evaluation. Communications, Signal Processing, and Systems : Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems. editor / Qilian Liang ; Wei Wang ; Xin Liu ; Zhenyu Na ; Baoju Zhang. Vol. 1 Springer Singapore, 2022. pp. 1004-1012 (Lecture Notes in Electrical Engineering).

@inproceedings{191ea110c8ac4027a92ef377926131fa,

title = "Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation",

abstract = "By grouping pixels with visual coherence, superpixel algorithms provide an alternative representation of regular pixel grid for precise and efficient image segmentation. In this paper, a multi-stage model is used for sea ice segmentation from the high-resolution optical imagery, including the pre-processing to enhance the image contrast and suppress the noise, superpixel generation and classification, and post-processing to refine the segmented results. Four superpixel algorithms are evaluated within the framework, where the high-resolution imagery of the Chukchi sea is used for validation. Quantitative evaluation in terms of the segmentation quality and floe size distribution, and visual comparison for several selected regions of interest are presented. Overall, the model with TS-SLIC yields the best results, with a segmentation accuracy of 98.19% on average and adhering to the ice edges well.",

keywords = "Satellite remote sensing, Sea ice segmentation, Superpixel",

author = "Siyuan Chen and Yijun Yan and Jinchang Ren and Byongjun Hwang and Stephen Marshall and Tariq Durrani",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 10th International Conference on Communications, Signal Processing, and Systems, CSPS 2021 ; Conference date: 24-07-2021 Through 25-07-2021",

year = "2022",

month = mar,

day = "31",

doi = "10.1007/978-981-19-0390-8_126",

language = "English",

isbn = "9789811903892",

volume = "1",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Singapore",

pages = "1004--1012",

editor = "Qilian Liang and Wei Wang and Xin Liu and Zhenyu Na and Baoju Zhang",

booktitle = "Communications, Signal Processing, and Systems",

address = "Singapore",

}

Chen, S, Yan, Y, Ren, J, Hwang, B, Marshall, S & Durrani, T 2022, Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation. in Q Liang, W Wang, X Liu, Z Na & B Zhang (eds), Communications, Signal Processing, and Systems : Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems. vol. 1, Lecture Notes in Electrical Engineering, vol. 878 LNEE, Springer Singapore, pp. 1004-1012, 10th International Conference on Communications, Signal Processing, and Systems, Changbaishan, China, 24/07/21. https://doi.org/10.1007/978-981-19-0390-8_126

Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images : Analysis and Evaluation. / Chen, Siyuan; Yan, Yijun; Ren, Jinchang et al.

Communications, Signal Processing, and Systems : Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems. ed. / Qilian Liang; Wei Wang; Xin Liu; Zhenyu Na; Baoju Zhang. Vol. 1 Springer Singapore, 2022. p. 1004-1012 (Lecture Notes in Electrical Engineering; Vol. 878 LNEE).

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

TY - GEN

T1 - Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images

T2 - 10th International Conference on Communications, Signal Processing, and Systems

AU - Chen, Siyuan

AU - Yan, Yijun

AU - Ren, Jinchang

AU - Hwang, Byongjun

AU - Marshall, Stephen

AU - Durrani, Tariq

N1 - Conference code: 10

PY - 2022/3/31

Y1 - 2022/3/31

N2 - By grouping pixels with visual coherence, superpixel algorithms provide an alternative representation of regular pixel grid for precise and efficient image segmentation. In this paper, a multi-stage model is used for sea ice segmentation from the high-resolution optical imagery, including the pre-processing to enhance the image contrast and suppress the noise, superpixel generation and classification, and post-processing to refine the segmented results. Four superpixel algorithms are evaluated within the framework, where the high-resolution imagery of the Chukchi sea is used for validation. Quantitative evaluation in terms of the segmentation quality and floe size distribution, and visual comparison for several selected regions of interest are presented. Overall, the model with TS-SLIC yields the best results, with a segmentation accuracy of 98.19% on average and adhering to the ice edges well.

AB - By grouping pixels with visual coherence, superpixel algorithms provide an alternative representation of regular pixel grid for precise and efficient image segmentation. In this paper, a multi-stage model is used for sea ice segmentation from the high-resolution optical imagery, including the pre-processing to enhance the image contrast and suppress the noise, superpixel generation and classification, and post-processing to refine the segmented results. Four superpixel algorithms are evaluated within the framework, where the high-resolution imagery of the Chukchi sea is used for validation. Quantitative evaluation in terms of the segmentation quality and floe size distribution, and visual comparison for several selected regions of interest are presented. Overall, the model with TS-SLIC yields the best results, with a segmentation accuracy of 98.19% on average and adhering to the ice edges well.

KW - Satellite remote sensing

KW - Sea ice segmentation

KW - Superpixel

UR - http://www.scopus.com/inward/record.url?scp=85128785136&partnerID=8YFLogxK

UR - https://link.springer.com/book/10.1007/978-981-19-0390-8

U2 - 10.1007/978-981-19-0390-8_126

DO - 10.1007/978-981-19-0390-8_126

M3 - Conference contribution

AN - SCOPUS:85128785136

SN - 9789811903892

SN - 9789811903922

VL - 1

T3 - Lecture Notes in Electrical Engineering

SP - 1004

EP - 1012

BT - Communications, Signal Processing, and Systems

A2 - Liang, Qilian

A2 - Wang, Wei

A2 - Liu, Xin

A2 - Na, Zhenyu

A2 - Zhang, Baoju

PB - Springer Singapore

Y2 - 24 July 2021 through 25 July 2021

ER -

Chen S, Yan Y, Ren J, Hwang B, Marshall S, Durrani T. Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation. In Liang Q, Wang W, Liu X, Na Z, Zhang B, editors, Communications, Signal Processing, and Systems : Proceedings of the 10th International Conference on Communications, Signal Processing, and Systems. Vol. 1. Springer Singapore. 2022. p. 1004-1012. (Lecture Notes in Electrical Engineering). Epub 2022 Mar 30. doi: 10.1007/978-981-19-0390-8_126

Superpixel Based Sea Ice Segmentation with High-Resolution Optical Images: Analysis and Evaluation

Abstract

Publication series

Conference

Access to Document

Other files and links

Cite this