A Synergistical Attention Model for Semantic Segmentation of Remote Sensing Images

Li, Xin; Xu, Feng; Liu, Fan; Lyu, Xin; Tong, Yao; Xu, Zhennan; Zhou, Jun

doi:10.1109/tgrs.2023.3243954

Cited by 32 publications

(17 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While various digital image processing methods have been proposed to improve the classification accuracy of detailed information in remote sensing images, they are often limited in their ability to meet the requirements of complex image classification [44][45][46]. Although machine learning methods have been shown to enhance classification accuracy [47][48][49][50], they require image preprocessing and feature enhancement to achieve better classification results [51]. Moreover, the degree of automation is not high, and significant improvements in the correct rate are already difficult to achieve.…”

Section: Coastal Wetland Detectionmentioning

confidence: 99%

Semantic Segmentation of China’s Coastal Wetlands Based on Sentinel-2 and Segformer

et al. 2023

View full text Add to dashboard Cite

Concerning the ever-changing wetland environment, the efficient extraction of wetland information holds great significance for the research and management of wetland ecosystems. China’s vast coastal wetlands possess rich and diverse geographical features. This study employs the SegFormer model and Sentinel-2 data to conduct a wetland classification study for coastal wetlands in Yancheng, Jiangsu, China. After preprocessing the Sentinel data, nine classification objects (construction land, Spartina alterniflora (S. alterniflora), Suaeda salsa (S. salsa), Phragmites australis (P. australis), farmland, river system, aquaculture and tidal falt) were identified based on the previous literature and remote sensing images. Moreover, mAcc, mIoU, aAcc, Precision, Recall and F-1 score were chosen as evaluation indicators. This study explores the potential and effectiveness of multiple methods, including data image processing, machine learning and deep learning. The results indicate that SegFormer is the best model for wetland classification, efficiently and accurately extracting small-scale features. With mIoU (0.81), mAcc (0.87), aAcc (0.94), mPrecision (0.901), mRecall (0.876) and mFscore (0.887) higher than other models. In the face of unbalanced wetland categories, combining CrossEntropyLoss and FocalLoss in the loss function can improve several indicators of difficult cases to be segmented, enhancing the classification accuracy and generalization ability of the model. Finally, the category scale pie chart of Yancheng Binhai wetlands was plotted. In conclusion, this study achieves an effective segmentation of Yancheng coastal wetlands based on the semantic segmentation method of deep learning, providing technical support and reference value for subsequent research on wetland values.

show abstract

Section: Coastal Wetland Detectionmentioning

confidence: 99%

Semantic Segmentation of China’s Coastal Wetlands Based on Sentinel-2 and Segformer

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Over the past years, many different attention modules have been designed for the semantic segmentation of RSIs [53,54]. DEANet extracts multiscale information by introducing pyramid sampling into the channel dimension [55].…”

Section: Attention Mechanismmentioning

confidence: 99%

MSAFNet: Multiscale Successive Attention Fusion Network for Water Body Extraction of Remote Sensing Images

Lyu

Jiang

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Water body extraction is a typical task in the semantic segmentation of remote sensing images (RSIs). Deep convolutional neural networks (DCNNs) outperform traditional methods in mining visual features; however, due to the inherent convolutional mechanism of the network, spatial details and abstract semantic representations at different levels are difficult to capture accurately at the same time, and then the extraction results decline to become suboptimal, especially on narrow areas and boundaries. To address the above-mentioned problem, a multiscale successive attention fusion network, named MSAFNet, is proposed to efficiently aggregate the multiscale features from two aspects. A successive attention fusion module (SAFM) is first devised to extract multiscale and fine-grained features of water bodies, while a joint attention module (JAM) is proposed to further mine salient semantic information by jointly modeling contextual dependencies. Furthermore, the multi-level features extracted by the above-mentioned modules are aggregated by a feature fusion module (FFM) so that the edges of water bodies are well mapped, directly improving the segmentation of various water bodies. Extensive experiments were conducted on the Qinghai-Tibet Plateau Lake (QTPL) and the Land-cOVEr Domain Adaptive semantic segmentation (LoveDA) datasets. Numerically, MSAFNet reached the highest accuracy on both QTPL and LoveDA datasets, including Kappa, MIoU, FWIoU, F1, and OA, outperforming several mainstream methods. Regarding the QTPL dataset, MSAFNet peaked at 99.14% and 98.97% in terms of F1 and OA. Although the LoveDA dataset is more challenging, MSAFNet retained the best performance, with F1 and OA being 97.69% and 95.87%. Additionally, visual inspections exhibited consistency with numerical evaluations.

show abstract

“…However, the aforementioned semantic segmentation models can only extract water features with a fixed receptive field size or extract multi-scale water body features with multiple receptive field sizes [34]- [37]. Nevertheless, the intricate semantics present in urban high-resolution remote sensing images often cause water features extracted based on local information to deviate from global information, thereby affecting the accuracy of urban water semantic segmentation [38].…”

Section: Introductionmentioning

confidence: 99%

AEDNet: An Attention-Based Encoder–Decoder Network for Urban Water Extraction From High Spatial Resolution Remote Sensing Images

Song,

Rui,

2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Accurate water extraction from urban remote sensing images holds great significance in assisting the formulation of river and lake management policies and ensuring the sustainable development of urban water resources. However, urban high-resolution remote sensing images encompass complex spatial and semantic information, which leads to disparities between the extracted water body features based on local and global information, consequently affecting the accuracy of urban water extraction. To tackle this issue, an attention-based encoder-decoder network (AEDNet) was proposed. In this network, the backbone employing atrous convolution facilitated the acquisition of low-level and high-level features of urban remote sensing images at various scales. Integrated with the attention mechanism, the encoderdecoder structure extracted global features in both the spatial and channel domains. Subsequently, these two types of features were merged to yield the urban water segmentation. Moreover, considering both intersection over union (IoU) and class weights, a joint loss function was introduced to further enhance the accuracy of urban water extraction. Experimental results demonstrated the strong performance of the proposed method on both GID and LoveDA datasets.

show abstract

A Synergistical Attention Model for Semantic Segmentation of Remote Sensing Images

Cited by 32 publications

References 69 publications

Semantic Segmentation of China’s Coastal Wetlands Based on Sentinel-2 and Segformer

Semantic Segmentation of China’s Coastal Wetlands Based on Sentinel-2 and Segformer

MSAFNet: Multiscale Successive Attention Fusion Network for Water Body Extraction of Remote Sensing Images

AEDNet: An Attention-Based Encoder–Decoder Network for Urban Water Extraction From High Spatial Resolution Remote Sensing Images

Contact Info

Product

Resources

About