Investigating the seasonal dynamics of surface water over the Qinghai–Tibet Plateau using Sentinel-1 imagery and a novel gated multiscale ConvNet

Luo, Xin; Hu, Zhiheng; Liu, Lin

doi:10.1080/17538947.2023.2198266

Cited by 6 publications

(3 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we define a standard convolution kernel R, denoted w as the weighted sum of sampling values, obtaining Equation (1), where R represents the regular grid of traditional convolution with a 3×3 convolution kernel of stride 1. Then, a standard convolution feature y(p_0 ) matrix can be obtained as shown in Equation (2). Finally, an offset ∆p_n is introduced into R to realize the offset of feature points.…”

Section: Deformable Convolutionmentioning

confidence: 99%

“…Surface water is an indispensable part of the Earth's ecosystem and a significant influencing factor in climate change, ecological environmental protection, and human production and life [1]. It plays crucial roles in various aspects such as environmental monitoring and management [2], ecological protection and restoration, disaster prevention and control, emergency response, agricultural production, land use, and people's water safety [3]. By obtaining and analyzing the spatial distribution and area information of water bodies, we can guide and support humans to adopt better ways of living [4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SwinDefNet: A Novel Surface Water Mapping Model in Mountain and Cloudy Regions Based on Sentinel-2 Imagery

Chen,

Pan,

Liu

2024

Preprint

View full text Add to dashboard Cite

Surface water plays a crucial role in climate change, human production, and life, making accurate monitoring and observation of surface water particularly important. However, due to the significant diversity and complexity of water distribution in surface space, accurate mapping of surface water faces considerable challenges. When extracting water bodies from medium-resolution satellite remote sensing images, CNN methods may suffer from limitations in receptive fields and insufficient context modeling capabilities, leading to the loss of water body boundary details and poor fusion of multiscale features. Currently, there is relatively little research on this issue; therefore, it is necessary to explore new combinations of deep learning networks to address these challenges. The purpose of this study is to address the above issues. We propose a new combination of deep learning networks that fully utilize multiscale information to enhance water features. Specifically, we first combine deformable convolutions with the Swin Transformer to increase effective receptive fields while better integrating global semantic information. This combination can capture features of water bodies at different scales, improve the accuracy and integrity of water extraction, and provide reliable technical support for detailed water body extraction. We tested the newly constructed model using Sentinel-2 satellite images. Our model achieved results of over 90%, with an average accuracy of 97.89%, average precision of 94.98%, average recall of 90.05%, and an average F1 score of 92.33%. In addition, our model achieved an accuracy of 98.03% in mountainous areas. Our experiments and results validate the potential of combining the Swin Transformer and deformable convolutions in detailed water body extraction.

show abstract

Section: Deformable Convolutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

SwinDefNet: A Novel Surface Water Mapping Model in Mountain and Cloudy Regions Based on Sentinel-2 Imagery

Chen,

Pan,

Liu

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The Qinghai-Tibet Plateau is known as the "Water Tower of Asia" due to its abundance of rivers, lakes, and glaciers [19][20][21], and climate change greatly affects the distribution of surface water in this region [22][23][24][25]. Luo et al [26] developed a new method for highprecision surface water mapping based on Sentinel-1 synthetic aperture radar (SAR) images and convolutional networks (ConvNets) in order to obtain information on the dynamics of surface water in the Qinghai-Tibet Plateau region. Guan [27] explored the evolution trend of surface water resources on the Qinghai-Tibet Plateau based on multi-period remote sensing data, and the results showed that the surface of the Qinghai-Tibet Plateau as a whole tends to be wetter, and the trend of wetter changes in the central and western regions of the plateau was more significant.…”

Section: Introductionmentioning

confidence: 99%

Characteristics and Influence Rules of Roadside Ponding along the Qinghai–Tibet Highway

Cui,

Zhu,

Liu

et al. 2024

Water

View full text Add to dashboard Cite

Due to climate change and seasonal precipitation, water conditions in the Qinghai–Tibet region are a significant factor affecting the stability of subgrades. The accumulation of large amounts of surface water leads to subgrade diseases along the Qinghai–Tibet Highway. Based on remote sensing photos obtained from Google Earth Engine and processing the photos using ENVI 5.6.3 and CAD 2019 software, this paper analyzed the distribution characteristics of surface water and studied the impact of roadside ponding on subgrade diseases. The results showed that the total area of surface water was more than 3.7 million m2, and the surface water was most widely distributed in large river areas such as the Tuotuo River and Buqu River. The subgrade diseases of the Qinghai–Tibet Highway could be categorized into three types: settlement, longitudinal crack, and frost boiling, which accounted for 71.09%, 17.13%, and 11.78% of the total number of subgrade diseases, respectively. Additionally, the ground mean annual temperature was an important factor affecting the distribution of surface water, with the surface water area showing an increasing trend with the increase in ground mean annual temperature, and roadside ponding was most likely to form in the high-temperature extremely unstable permafrost area.

show abstract

GDSNet: A gated dual-stream convolutional neural network for automatic recognition of coseismic landslides

Wang,

Zheng

et al. 2024

International Journal of Applied Earth Observation and Geoinfor

View full text Add to dashboard Cite

Investigating the seasonal dynamics of surface water over the Qinghai–Tibet Plateau using Sentinel-1 imagery and a novel gated multiscale ConvNet

Cited by 6 publications

References 58 publications

SwinDefNet: A Novel Surface Water Mapping Model in Mountain and Cloudy Regions Based on Sentinel-2 Imagery

SwinDefNet: A Novel Surface Water Mapping Model in Mountain and Cloudy Regions Based on Sentinel-2 Imagery

Characteristics and Influence Rules of Roadside Ponding along the Qinghai–Tibet Highway

GDSNet: A gated dual-stream convolutional neural network for automatic recognition of coseismic landslides

Contact Info

Product

Resources

About