Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images

Xudong Kang; Yufan Huang; Shutao Li; Hui Lin; Jon Atli Benediktsson

doi:10.1109/TGRS.2017.2755773

Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images

Xudong Kang, Yufan Huang, Shutao Li^*, Hui Lin, Jon Atli Benediktsson

^*Corresponding author for this work

University of Iceland

Research output: Contribution to journal › Article › peer-review

34 Citations (Scopus)

Abstract

The existence of shadows in very high resolution satellite images obstructs image interpretation and the following applications, such as target detection and recognition. Traditional shadow detection methods consider only the pixel-level properties, such as color and intensity of image pixels, and thus, may produce errors around object boundaries. To overcome this problem, a novel shadow detection algorithm based on extended random walker (ERW) is proposed by jointly integrating both shadow property and spatial correlations among adjacent pixels. First, a set of training samples is automatically generated via an improved Otsu-based thresholding method. Then, the support vector machine is applied to obtain an initial detection map, which categorizes all the pixels in the scene into shadow and nonshadow. Finally, the initial detection map is refined with the ERW model, which can simultaneously characterize the shadow property and spatial information in satellite images to further improve shadow detection accuracy. Experiments performed on five real remote sensing images demonstrate the superiority of the proposed method over several state-of-the-art methods in terms of detection accuracy.

Original language	English
Article number	8067651
Pages (from-to)	867-876
Number of pages	10
Journal	IEEE Transactions on Geoscience and Remote Sensing
Volume	56
Issue number	2
DOIs	https://doi.org/10.1109/TGRS.2017.2755773
Publication status	Published - Feb 2018

Bibliographical note

Funding Information:
Manuscript received May 4, 2017; revised July 21, 2017; accepted September 11, 2017. Date of publication October 13, 2017; date of current version January 26, 2018. This work was supported in part by the National Natural Science Fund of China for International Cooperation and Exchanges under Grant 61520106001, in part by the National Natural Science Foundation of China under Grant 61601179, and in part by the National Natural Science Fund of China for Distinguished Young Scholars under Grant 61325007. (Corresponding author: Shutao Li.) X. Kang, Y. Huang, S. Li, and H. Lin are with the College of Electrical and Information Engineering, Hunan University, Changsha 410082, China (e-mail: [email protected]; [email protected]; [email protected]; [email protected]).

Publisher Copyright:
© 1980-2012 IEEE.

Other keywords

Extended random walker (ERW)
remote sensing image
shadow detection
support vector machine (SVM)

Access to Document

10.1109/TGRS.2017.2755773

Cite this

@article{9b63d56a3d534745afba16a7dd145fb3,

title = "Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images",

abstract = "The existence of shadows in very high resolution satellite images obstructs image interpretation and the following applications, such as target detection and recognition. Traditional shadow detection methods consider only the pixel-level properties, such as color and intensity of image pixels, and thus, may produce errors around object boundaries. To overcome this problem, a novel shadow detection algorithm based on extended random walker (ERW) is proposed by jointly integrating both shadow property and spatial correlations among adjacent pixels. First, a set of training samples is automatically generated via an improved Otsu-based thresholding method. Then, the support vector machine is applied to obtain an initial detection map, which categorizes all the pixels in the scene into shadow and nonshadow. Finally, the initial detection map is refined with the ERW model, which can simultaneously characterize the shadow property and spatial information in satellite images to further improve shadow detection accuracy. Experiments performed on five real remote sensing images demonstrate the superiority of the proposed method over several state-of-the-art methods in terms of detection accuracy.",

keywords = "Extended random walker (ERW), remote sensing image, shadow detection, support vector machine (SVM)",

author = "Xudong Kang and Yufan Huang and Shutao Li and Hui Lin and Benediktsson, {Jon Atli}",

note = "Funding Information: Manuscript received May 4, 2017; revised July 21, 2017; accepted September 11, 2017. Date of publication October 13, 2017; date of current version January 26, 2018. This work was supported in part by the National Natural Science Fund of China for International Cooperation and Exchanges under Grant 61520106001, in part by the National Natural Science Foundation of China under Grant 61601179, and in part by the National Natural Science Fund of China for Distinguished Young Scholars under Grant 61325007. (Corresponding author: Shutao Li.) X. Kang, Y. Huang, S. Li, and H. Lin are with the College of Electrical and Information Engineering, Hunan University, Changsha 410082, China (e-mail: [email protected]; [email protected]; [email protected]; [email protected]). Publisher Copyright: {\textcopyright} 1980-2012 IEEE.",

year = "2018",

month = feb,

doi = "10.1109/TGRS.2017.2755773",

language = "English",

volume = "56",

pages = "867--876",

journal = "IEEE Transactions on Geoscience and Remote Sensing",

issn = "0196-2892",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "2",

}

TY - JOUR

T1 - Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images

AU - Kang, Xudong

AU - Huang, Yufan

AU - Li, Shutao

AU - Lin, Hui

AU - Benediktsson, Jon Atli

N1 - Funding Information: Manuscript received May 4, 2017; revised July 21, 2017; accepted September 11, 2017. Date of publication October 13, 2017; date of current version January 26, 2018. This work was supported in part by the National Natural Science Fund of China for International Cooperation and Exchanges under Grant 61520106001, in part by the National Natural Science Foundation of China under Grant 61601179, and in part by the National Natural Science Fund of China for Distinguished Young Scholars under Grant 61325007. (Corresponding author: Shutao Li.) X. Kang, Y. Huang, S. Li, and H. Lin are with the College of Electrical and Information Engineering, Hunan University, Changsha 410082, China (e-mail: [email protected]; [email protected]; [email protected]; [email protected]). Publisher Copyright: © 1980-2012 IEEE.

PY - 2018/2

Y1 - 2018/2

N2 - The existence of shadows in very high resolution satellite images obstructs image interpretation and the following applications, such as target detection and recognition. Traditional shadow detection methods consider only the pixel-level properties, such as color and intensity of image pixels, and thus, may produce errors around object boundaries. To overcome this problem, a novel shadow detection algorithm based on extended random walker (ERW) is proposed by jointly integrating both shadow property and spatial correlations among adjacent pixels. First, a set of training samples is automatically generated via an improved Otsu-based thresholding method. Then, the support vector machine is applied to obtain an initial detection map, which categorizes all the pixels in the scene into shadow and nonshadow. Finally, the initial detection map is refined with the ERW model, which can simultaneously characterize the shadow property and spatial information in satellite images to further improve shadow detection accuracy. Experiments performed on five real remote sensing images demonstrate the superiority of the proposed method over several state-of-the-art methods in terms of detection accuracy.

AB - The existence of shadows in very high resolution satellite images obstructs image interpretation and the following applications, such as target detection and recognition. Traditional shadow detection methods consider only the pixel-level properties, such as color and intensity of image pixels, and thus, may produce errors around object boundaries. To overcome this problem, a novel shadow detection algorithm based on extended random walker (ERW) is proposed by jointly integrating both shadow property and spatial correlations among adjacent pixels. First, a set of training samples is automatically generated via an improved Otsu-based thresholding method. Then, the support vector machine is applied to obtain an initial detection map, which categorizes all the pixels in the scene into shadow and nonshadow. Finally, the initial detection map is refined with the ERW model, which can simultaneously characterize the shadow property and spatial information in satellite images to further improve shadow detection accuracy. Experiments performed on five real remote sensing images demonstrate the superiority of the proposed method over several state-of-the-art methods in terms of detection accuracy.

KW - Extended random walker (ERW)

KW - remote sensing image

KW - shadow detection

KW - support vector machine (SVM)

UR - http://www.scopus.com/inward/record.url?scp=85041320639&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2017.2755773

DO - 10.1109/TGRS.2017.2755773

M3 - Article

AN - SCOPUS:85041320639

SN - 0196-2892

VL - 56

SP - 867

EP - 876

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

IS - 2

M1 - 8067651

ER -

Extended Random Walker for Shadow Detection in Very High Resolution Remote Sensing Images

Abstract

Bibliographical note

Other keywords

Access to Document

Fingerprint

Cite this