A Semantic Segmentation Method of Remote Sensing Image Based on Feature Fusion and Attention Mechanism
3
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Current methods for semantic segmentation of remotesensing images, especially for irregular and small targets,often result in low precision and incomplete feature extraction. To address this issue, an improved semanticsegmentation method was developed utilizing DeepLabv3+. First, DeepLabv3+ is combined with the proposedfeature fusion module to make full use of the complementary information of low and highlevel features. Second, the channel attention module helps extract effective features while suppressing irrelevant features,thereby enabling the extraction of more meaningful global information from highlevel features. Finally, richspatial information is selected using guided spatial attention, which improves the accuracy of edge segmentationof target objects. The results of the comparison show that the mean F1 score (MF1) and overall accuracy (OA)of the proposed method on the ISPRS Potsdam dataset are 89.81% and 88.45%, respectively. The MF1 of theproposed method is 89.90% and the OA is 89.14% for the UAVid dataset, which are higher than those of theother comparison algorithms. The proposed method exhibits superior semantic segmentation capabilities forremotesensing images.
Abstract
Current methods for semantic segmentation of remotesensing images, especially for irregular and small targets,often result in low precision and incomplete feature extraction. To address this issue, an improved semanticsegmentation method was developed utilizing DeepLabv3+. First, DeepLabv3+ is combined with the proposedfeature fusion module to make full use of the complementary information of low and highlevel features. Second, the channel attention module helps extract effective features while suppressing irrelevant features,thereby enabling the extraction of more meaningful global information from highlevel features. Finally, richspatial information is selected using guided spatial attention, which improves the accuracy of edge segmentationof target objects. The results of the comparison show that the mean F1 score (MF1) and overall accuracy (OA)of the proposed method on the ISPRS Potsdam dataset are 89.81% and 88.45%, respectively. The MF1 of theproposed method is 89.90% and the OA is 89.14% for the UAVid dataset, which are higher than those of theother comparison algorithms. The proposed method exhibits superior semantic segmentation capabilities forremotesensing images.
참고문헌 (30)
- B. Cui, "Semantic segmentation of remote sensing images using transfer learning and deep convolutional neural network with dense connection", IEEE Access, 8, pp.116744~116755 (2020)[DOI]
- W. Sun, "Fully convolutional networks for semantic segmentation of very high resolution remotely sensed images combined with DSM", IEEE Geoscience and Remote Sensing Letters, 15(3), pp.474~478 (2018)[DOI]
- J. Jiang, "RWSNet : a semantic segmentation network based on SegNet combined with random walk for remote sensing", International Journal of Remote Sensing, 41(2), pp.487~505 (2020)[DOI]
- L. Ding, "LANet : Local attention embedding to improve the semantic segmentation of remote sensing images", IEEE Transactions on Geoscience and Remote Sensing, 59(1), pp.426~435 (2021)[DOI]
- J. Zhang, "Multi-scale context aggregation for semantic segmentation of remote sensing images", Remote Sensing, 12(4) (2020)[DOI]
- M. Kampffmeyer, "Semantic segmentation of small objects and modeling of uncertainty in urban remote sensing images using deep convolutional neural networks", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp.1~9 (2016)
- X. Yuan, "A review of deep learning methods for semantic segmentation of remote sensing imagery", Expert Systems with Applications, 169 (2021)[DOI]
- R. Dong, "DenseU-net-based semantic segmentation of small objects in urban remote sensing images", IEEE Access, 7, pp.65347~65356 (2019)[DOI]
- R. Shang, "Multi-scale adaptive feature fusion network for semantic segmentation in remote sensing images", Remote Sensing, 12(5) (2020)[DOI]
- L. Mi, "Superpixel-enhanced deep neural forest for remote sensing image semantic segmentation", ISPRS Journal of Photogrammetry and Remote Sensing, 159, pp.140~152 (2020)[DOI]
- Y. Hua, "Semantic segmentation of remote sensing images with sparse annotations", IEEE Geoscience and Remote Sensing Letters, 19 (2021)[DOI]
- Z. Xu, "HRCNet : High-resolution context extraction network for semantic segmentation of remote sensing images", Remote Sensing, 13(1) (2020)[DOI]
- H. Wei, "DEANet : dual encoder with attention network for semantic segmentation of remote sensing imagery", Remote Sensing, 13(19) (2021)[DOI]
- Y. Liu, "Efficient patch-wise semantic segmentation for large-scale remote sensing images", Sensors, 18(10) (2018)[DOI]
- M. Alam, "Convolutional neural network for the semantic segmentation of remote sensing images", Mobile Networks and Applications, 26, pp.200~215 (2021)[DOI]
- C. He, "Remote sensing image semantic segmentation based on edge information guidance", Remote Sensing, 12(9) (2020)[DOI]
- R. Kemker, "Algorithms for semantic segmentation of multispectral remote sensing imagery using deep learning", ISPRS Journal of Photogrammetry and Remote Sensing, 145, pp.60~77 (2018)[DOI]
- Y. Li, "Learning deep semantic segmentation network under multiple weakly-supervised constraints for cross-domain remote sensing image semantic segmentation", ISPRS Journal of Photogrammetry and Remote Sensing, 175, pp.20~33 (2021)[DOI]
- R. Li, "Multiattention network for semantic segmentation of fine-resolution remote sensing images", IEEE Transactions on Geoscience and Remote Sensing, 60 (2021)[DOI]
- S. Ouyang, "Combining deep semantic segmentation network and graph convolutional neural network for semantic segmentation of remote sensing imagery", Remote Sensing, 13(1) (2020)[DOI]
- Z. Zhang, "Road extraction by deep residual U-Net", IEEE Geoscience and Remote Sensing Letters, 15(5), pp.749~753 (2018)[DOI]
- H. Yang, "Building extraction in very high resolution imagery by dense-attention networks", Remote Sensing, 10(11) (2018)[DOI]
- Y. Yi, "Semantic segmentation of urban buildings from VHR remote sensing imagery using a deep convolutional neural network", Remote Sensing, 11(15) (2019)[DOI]
- F. I. Diakogiannis, "ResUNet-a : a deep learning framework for semantic segmentation of remotely sensed data", ISPRS Journal of Photogrammetry and Remote Sensing, 162, pp.94~114 (2020)[DOI]
- X. Pan, "High-resolution aerial imagery semantic labeling with dense pyramid network", Sensors, 18(11) (2018)[DOI]
- R. Niu, "Hybrid multiple attention network for semantic segmentation in aerial images", IEEE Transactions on Geoscience and Remote Sensing, 60 (2021)[DOI]
- L. C. Chen, "DeepLab : semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs", IEEE Transactions on Pattern Analysis and Machine Intelligence, 40(4), pp.834~848 (2017)[DOI]
- L. Yuan, "Remote Sensing Image Classification Based on DeepLab-v3+", Laser &Optoelectronics Progress, 56(15) (2019)[DOI]
- B. Chen, "Semantic image segmentation network based on deep learning", Proceedings of SPIE 11429: MIPPR 2019: Automatic Target Recognition and Navigation, pp.77~81 (2020), International Society for Optics and Photonics
- L. Tian, "Semantic segmentation of remote sensing image based on GAN and FCN network model", Scientific Programming, 2021 (2021)[DOI]