Water Body Automated Extraction in Polarization SAR Images With Dense-Coordinate-Feature-Concatenate Network

Xue, Weibao; Yang, Hui; Wu, Yue; Kong, Ping; Xu, Hao; Wu, Penghai; Ma, Xiaoshuang

doi:10.1109/jstars.2021.3129182

Cited by 12 publications

(3 citation statements)

References 56 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of non-local speckle filters for coastline extraction purposes is addressed in [19], where single-polarization X-band TanDEM-X pursuit monostatic SAR measurements were considered. In [20], the "twin-problem" related to the extraction of the waterline associated with a reservoir is addressed using dual-polarimetric SAR measurements with the main goal of estimating the waterbody area. The accuracy of the estimated waterbody area is contrasted with estimates obtained using single-polarization SAR measurements and with ancillary in situ observation.…”

Section: Introductionmentioning

confidence: 99%

The Use of Satellite Synthetic Aperture Radar Imagery to Assist in the Monitoring of the Time Evolution of Challenging Coastal Environments: A Case Study of the Basilicata Coast

Ferrentino,

Famiglietti,

Nunziata

et al. 2023

Environments

View full text Add to dashboard Cite

This study focuses on a very complex environment, namely the Ionian coast of the Basilicata region, Southern Italy, which includes different kinds of beaches, river mouths and built-up areas. This complex environment is used as a test case to analyze the time variability of the coastline using measurements that were remotely sensed by the satellite European Copernicus Synthetic Aperture Radar (SAR) mission. First, the accuracy of the coastline, extracted by the SAR, is discussed with respect to finer-spatial-resolution drone-based light detection and ranging (LIDAR) measurements. Then, a time series of SAR dual-polarimetric measurements acquired by the European Copernicus mission is used to discuss the time variability of the coastline of the area of interest in a time period spanning from 2015 to 2021. The experimental results show that the accuracy of the SAR-based coastline is better than 15 m, which is reasonably good precision for monitoring the erosion/accretion processes that characterize the area of interest at a moderate scale. The estimated time variability of the extracted coastline suggests a dominant erosion process, which is always within 60 m.

show abstract

Section: Introductionmentioning

confidence: 99%

The Use of Satellite Synthetic Aperture Radar Imagery to Assist in the Monitoring of the Time Evolution of Challenging Coastal Environments: A Case Study of the Basilicata Coast

Ferrentino,

Famiglietti,

Nunziata

et al. 2023

Environments

View full text Add to dashboard Cite

show abstract

“…In the work by Denbina et al [18], CNNs are adopted to detect flooding in urban areas, which suggests the potential of CNN-based urban water extraction in SAR images. Furthermore, a dense-coordinate-feature-concatenate network (DCFNet) is proposed to extract and fuse the water features [19]. The experimental results on Gaofen-3 and Sentinel-1 SAR images show that DCFNet can reduce the influence of ground interference and speckle noise to some extent.…”

Section: Introductionmentioning

confidence: 99%

HA-Unet: A Modified Unet Based on Hybrid Attention for Urban Water Extraction in SAR Images

Song

Huang

et al. 2022

Electronics

View full text Add to dashboard Cite

Urban water plays a significant role in the urban ecosystem, but urban water extraction is still a challenging task in automatic interpretation of synthetic aperture radar (SAR) images. The influence of radar shadows and strong scatters in urban areas may lead to misclassification in urban water extraction. Nevertheless, the local features captured by convolutional layers in Convolutional Neural Networks (CNNs) are generally redundant and cannot make effective use of global information to guide the prediction of water pixels. To effectively emphasize the identifiable water characteristics and fully exploit the global information of SAR images, a modified Unet based on hybrid attention mechanism is proposed to improve the performance of urban water extraction in this paper. Considering the feature extraction ability and the global modeling capability in SAR image segmentation, the Channel and Spatial Attention Module (CSAM) and the Multi-head Self-Attention Block (MSAB) are both introduced into the proposed Hybrid Attention Unet (HA-Unet). In this work, Resnet50 is adopted as the backbone of HA-Unet to extract multi-level features of SAR images. During the feature extraction process, CSAM based on local attention is adopted to enhance the meaningful water features and ignore unnecessary features adaptively in feature maps of two shallow layers. In the last two layers of the backbone, MSAB is introduced to capture the global information of SAR images to generate global attention. In addition, two global attention maps generated by MSAB are aggregated together to reconstruct the spatial feature relationship of SAR images from high-resolution feature maps. The experimental results on Sentinel-1A SAR images show that the proposed urban water extraction method has a strong ability to extract water bodies in the complex urban areas. The ablation experiment and visualization results vividly indicate that both CSAM and MSAB contribute significantly to extracting urban water accurately and effectively.

show abstract

“…However, multi-polarization SAR images can provide more information about water body properties. Furthermore, rough water surfaces may be blanketed in copolarization SAR data, whereas cross-polarization SAR data can still provide water evidence [25]. In [17], H-A-α polarimetric decomposition is integrated with FCN to exploit the polarimetric information in PolSAR images, with an overall accuracy rate of up to 95%.…”

Section: Introductionmentioning

confidence: 99%

Flood Detection in Dual-Polarization SAR Images Based on Multi-Scale Deeplab Model

Song

Huang

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

The proliferation of massive polarimetric Synthetic Aperture Radar (SAR) data helps promote the development of SAR image interpretation. Due to the advantages of powerful feature extraction capability and strong adaptability for different tasks, deep learning has been adopted in the work of SAR image interpretation and has achieved good results. However, most deep learning methods only employ single-polarization SAR images and ignore the water features embedded in multi-polarization SAR images. To fully exploit the dual-polarization SAR data and multi-scale features of SAR images, an effective flood detection method for SAR images is proposed in this paper. In the proposed flood detection method, a powerful Multi-Scale Deeplab (MS-Deeplab) model is constructed based on the dual-channel MobileNetV2 backbone and the classic DeeplabV3+ architecture to improve the ability of water feature extraction in SAR images. Firstly, the dual-channel feature extraction backbone based on the lightweight MobileNetV2 separately trains the dual-polarization SAR images, and the obtained training parameters are merged with the linear weighting to fuse dual-polarization water features. Given the multi-scale space information in SAR images, then, a multi-scale feature fusion module is introduced to effectively utilize multi-layer features and contextual information, which enhances the representation of water features. Finally, a joint loss function is constructed based on cross-entropy and a dice coefficient to deal with the imbalanced categorical distribution in the training dataset. The experimental results on the time series of Sentinel-1A SAR images show that the proposed method for flood detection has a strong ability to locate water boundaries and tiny water bodies in complex scenes. In terms of quantitative assessment, MS-Deeplab can achieve a better performance compared with other mainstream semantic segmentation models, including PSPNet, Unet and the original DeeplabV3+ model, with a 3.27% intersection over union (IoU) and 1.69% pixel accuracy (PA) improvement than the original DeeplabV3+ model.

show abstract

Water Body Automated Extraction in Polarization SAR Images With Dense-Coordinate-Feature-Concatenate Network

Cited by 12 publications

References 56 publications

The Use of Satellite Synthetic Aperture Radar Imagery to Assist in the Monitoring of the Time Evolution of Challenging Coastal Environments: A Case Study of the Basilicata Coast

The Use of Satellite Synthetic Aperture Radar Imagery to Assist in the Monitoring of the Time Evolution of Challenging Coastal Environments: A Case Study of the Basilicata Coast

HA-Unet: A Modified Unet Based on Hybrid Attention for Urban Water Extraction in SAR Images

Flood Detection in Dual-Polarization SAR Images Based on Multi-Scale Deeplab Model

Contact Info

Product

Resources

About