Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor

Genyun Sun, Yanling Hao, Xiaolin Chen, Jinchang Ren, Aizhu Zhang, Binghu Huang, Yuanzhi Zhang, Xiuping Jia

Research output: Contribution to journalArticle

  • 2 Citations

Abstract

The region merging algorithm is a widely used segmentation technique for very high resolution (VHR) remote sensing images. However, the segmentation of post-earthquake VHR images is more difficult due to the complexity of these images, especially high intra-class and low inter-class variability among damage objects. Herein two key issues must be resolved: the first is to find an appropriate descriptor to measure the similarity of two adjacent regions since they exhibit high complexity among the diverse damage objects, such as landslides, debris flow, and collapsed buildings. The other is how to solve over-segmentation and under-segmentation problems, which are commonly encountered with conventional merging strategies due to their strong dependence on local information. To tackle these two issues, an adaptive dynamic region merging approach (ADRM) is introduced, which combines an adaptive spectral-spatial descriptor and a dynamic merging strategy to adapt to the changes of merging regions for successfully detecting objects scattered globally in a post-earthquake image. In the new descriptor, the spectral similarity and spatial similarity of any two adjacent regions are automatically combined to measure their similarity. Accordingly, the new descriptor offers adaptive semantic descriptions for geo-objects and thus is capable of characterizing different damage objects. Besides, in the dynamic region merging strategy, the adaptive spectral-spatial descriptor is embedded in the defined testing order and combined with graph models to construct a dynamic merging strategy. The new strategy can find the global optimal merging order and ensures that the most similar regions are merged at first. With combination of the two strategies, ADRM can identify spatially scattered objects and alleviates the phenomenon of over-segmentation and under-segmentation. The performance of ADRM has been evaluated by comparing with four state-of-the-art segmentation methods, including the fractal net evolution approach (FNEA, as implemented in the eCognition software, Trimble Inc., Westminster, CO, USA), the J-value segmentation (JSEG) method, the graph-based segmentation (GSEG) method, and the statistical region merging (SRM) approach. The experiments were conducted on six VHR subarea images captured by RGB sensors mounted on aerial platforms, which were acquired after the 2008 Wenchuan Ms 8.0 earthquake. Quantitative and qualitative assessments demonstrated that the proposed method offers high feasibility and improved accuracy in the segmentation of post-earthquake VHR aerial images.
LanguageEnglish
Article number899
Number of pages24
JournalRemote Sensing
Volume9
Issue number9
DOIs
StatePublished - 30 Aug 2017

Fingerprint

segmentation
earthquake
image resolution
damage
debris flow
landslide
sensor
remote sensing
software
method

Keywords

  • post-earthquake image segmentation
  • adaptive spectral-spatial descriptor
  • dynamic region merging

Cite this

Sun, G., Hao, Y., Chen, X., Ren, J., Zhang, A., Huang, B., ... Jia, X. (2017). Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor. Remote Sensing, 9(9), [899]. DOI: 10.3390/rs9090899
Sun, Genyun ; Hao, Yanling ; Chen, Xiaolin ; Ren, Jinchang ; Zhang, Aizhu ; Huang, Binghu ; Zhang, Yuanzhi ; Jia, Xiuping. / Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor. In: Remote Sensing. 2017 ; Vol. 9, No. 9.
@article{7a689799fb584f7f884bd0f0ab354ad1,
title = "Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor",
abstract = "The region merging algorithm is a widely used segmentation technique for very high resolution (VHR) remote sensing images. However, the segmentation of post-earthquake VHR images is more difficult due to the complexity of these images, especially high intra-class and low inter-class variability among damage objects. Herein two key issues must be resolved: the first is to find an appropriate descriptor to measure the similarity of two adjacent regions since they exhibit high complexity among the diverse damage objects, such as landslides, debris flow, and collapsed buildings. The other is how to solve over-segmentation and under-segmentation problems, which are commonly encountered with conventional merging strategies due to their strong dependence on local information. To tackle these two issues, an adaptive dynamic region merging approach (ADRM) is introduced, which combines an adaptive spectral-spatial descriptor and a dynamic merging strategy to adapt to the changes of merging regions for successfully detecting objects scattered globally in a post-earthquake image. In the new descriptor, the spectral similarity and spatial similarity of any two adjacent regions are automatically combined to measure their similarity. Accordingly, the new descriptor offers adaptive semantic descriptions for geo-objects and thus is capable of characterizing different damage objects. Besides, in the dynamic region merging strategy, the adaptive spectral-spatial descriptor is embedded in the defined testing order and combined with graph models to construct a dynamic merging strategy. The new strategy can find the global optimal merging order and ensures that the most similar regions are merged at first. With combination of the two strategies, ADRM can identify spatially scattered objects and alleviates the phenomenon of over-segmentation and under-segmentation. The performance of ADRM has been evaluated by comparing with four state-of-the-art segmentation methods, including the fractal net evolution approach (FNEA, as implemented in the eCognition software, Trimble Inc., Westminster, CO, USA), the J-value segmentation (JSEG) method, the graph-based segmentation (GSEG) method, and the statistical region merging (SRM) approach. The experiments were conducted on six VHR subarea images captured by RGB sensors mounted on aerial platforms, which were acquired after the 2008 Wenchuan Ms 8.0 earthquake. Quantitative and qualitative assessments demonstrated that the proposed method offers high feasibility and improved accuracy in the segmentation of post-earthquake VHR aerial images.",
keywords = "post-earthquake image segmentation, adaptive spectral-spatial descriptor, dynamic region merging",
author = "Genyun Sun and Yanling Hao and Xiaolin Chen and Jinchang Ren and Aizhu Zhang and Binghu Huang and Yuanzhi Zhang and Xiuping Jia",
year = "2017",
month = "8",
day = "30",
doi = "10.3390/rs9090899",
language = "English",
volume = "9",
journal = "Remote Sensing",
issn = "2072-4292",
number = "9",

}

Sun, G, Hao, Y, Chen, X, Ren, J, Zhang, A, Huang, B, Zhang, Y & Jia, X 2017, 'Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor' Remote Sensing, vol. 9, no. 9, 899. DOI: 10.3390/rs9090899

Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor. / Sun, Genyun; Hao, Yanling; Chen, Xiaolin; Ren, Jinchang; Zhang, Aizhu; Huang, Binghu; Zhang, Yuanzhi; Jia, Xiuping.

In: Remote Sensing, Vol. 9, No. 9, 899, 30.08.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor

AU - Sun,Genyun

AU - Hao,Yanling

AU - Chen,Xiaolin

AU - Ren,Jinchang

AU - Zhang,Aizhu

AU - Huang,Binghu

AU - Zhang,Yuanzhi

AU - Jia,Xiuping

PY - 2017/8/30

Y1 - 2017/8/30

N2 - The region merging algorithm is a widely used segmentation technique for very high resolution (VHR) remote sensing images. However, the segmentation of post-earthquake VHR images is more difficult due to the complexity of these images, especially high intra-class and low inter-class variability among damage objects. Herein two key issues must be resolved: the first is to find an appropriate descriptor to measure the similarity of two adjacent regions since they exhibit high complexity among the diverse damage objects, such as landslides, debris flow, and collapsed buildings. The other is how to solve over-segmentation and under-segmentation problems, which are commonly encountered with conventional merging strategies due to their strong dependence on local information. To tackle these two issues, an adaptive dynamic region merging approach (ADRM) is introduced, which combines an adaptive spectral-spatial descriptor and a dynamic merging strategy to adapt to the changes of merging regions for successfully detecting objects scattered globally in a post-earthquake image. In the new descriptor, the spectral similarity and spatial similarity of any two adjacent regions are automatically combined to measure their similarity. Accordingly, the new descriptor offers adaptive semantic descriptions for geo-objects and thus is capable of characterizing different damage objects. Besides, in the dynamic region merging strategy, the adaptive spectral-spatial descriptor is embedded in the defined testing order and combined with graph models to construct a dynamic merging strategy. The new strategy can find the global optimal merging order and ensures that the most similar regions are merged at first. With combination of the two strategies, ADRM can identify spatially scattered objects and alleviates the phenomenon of over-segmentation and under-segmentation. The performance of ADRM has been evaluated by comparing with four state-of-the-art segmentation methods, including the fractal net evolution approach (FNEA, as implemented in the eCognition software, Trimble Inc., Westminster, CO, USA), the J-value segmentation (JSEG) method, the graph-based segmentation (GSEG) method, and the statistical region merging (SRM) approach. The experiments were conducted on six VHR subarea images captured by RGB sensors mounted on aerial platforms, which were acquired after the 2008 Wenchuan Ms 8.0 earthquake. Quantitative and qualitative assessments demonstrated that the proposed method offers high feasibility and improved accuracy in the segmentation of post-earthquake VHR aerial images.

AB - The region merging algorithm is a widely used segmentation technique for very high resolution (VHR) remote sensing images. However, the segmentation of post-earthquake VHR images is more difficult due to the complexity of these images, especially high intra-class and low inter-class variability among damage objects. Herein two key issues must be resolved: the first is to find an appropriate descriptor to measure the similarity of two adjacent regions since they exhibit high complexity among the diverse damage objects, such as landslides, debris flow, and collapsed buildings. The other is how to solve over-segmentation and under-segmentation problems, which are commonly encountered with conventional merging strategies due to their strong dependence on local information. To tackle these two issues, an adaptive dynamic region merging approach (ADRM) is introduced, which combines an adaptive spectral-spatial descriptor and a dynamic merging strategy to adapt to the changes of merging regions for successfully detecting objects scattered globally in a post-earthquake image. In the new descriptor, the spectral similarity and spatial similarity of any two adjacent regions are automatically combined to measure their similarity. Accordingly, the new descriptor offers adaptive semantic descriptions for geo-objects and thus is capable of characterizing different damage objects. Besides, in the dynamic region merging strategy, the adaptive spectral-spatial descriptor is embedded in the defined testing order and combined with graph models to construct a dynamic merging strategy. The new strategy can find the global optimal merging order and ensures that the most similar regions are merged at first. With combination of the two strategies, ADRM can identify spatially scattered objects and alleviates the phenomenon of over-segmentation and under-segmentation. The performance of ADRM has been evaluated by comparing with four state-of-the-art segmentation methods, including the fractal net evolution approach (FNEA, as implemented in the eCognition software, Trimble Inc., Westminster, CO, USA), the J-value segmentation (JSEG) method, the graph-based segmentation (GSEG) method, and the statistical region merging (SRM) approach. The experiments were conducted on six VHR subarea images captured by RGB sensors mounted on aerial platforms, which were acquired after the 2008 Wenchuan Ms 8.0 earthquake. Quantitative and qualitative assessments demonstrated that the proposed method offers high feasibility and improved accuracy in the segmentation of post-earthquake VHR aerial images.

KW - post-earthquake image segmentation

KW - adaptive spectral-spatial descriptor

KW - dynamic region merging

U2 - 10.3390/rs9090899

DO - 10.3390/rs9090899

M3 - Article

VL - 9

JO - Remote Sensing

T2 - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 9

M1 - 899

ER -

Sun G, Hao Y, Chen X, Ren J, Zhang A, Huang B et al. Dynamic post-earthquake image segmentation with an adaptive spectral-spatial descriptor. Remote Sensing. 2017 Aug 30;9(9). 899. Available from, DOI: 10.3390/rs9090899