PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module
In the domain of remote sensing research, the extraction of roads from high-resolution imagery remains a formidable challenge. In this paper, we introduce an advanced architecture called PCCAU-Net, which integrates Pyramid Pathway Input, CoordConv convolution, and Dual-Inut Cross Attention (DCA) mod...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/4/1634 |
| _version_ | 1797298961714774016 |
|---|---|
| author | Xiaoqin Xue Chao Ren Anchao Yin Ying Zhou Yuanyuan Liu Cong Ding Jiakai Lu |
| author_facet | Xiaoqin Xue Chao Ren Anchao Yin Ying Zhou Yuanyuan Liu Cong Ding Jiakai Lu |
| author_sort | Xiaoqin Xue |
| collection | DOAJ |
| description | In the domain of remote sensing research, the extraction of roads from high-resolution imagery remains a formidable challenge. In this paper, we introduce an advanced architecture called PCCAU-Net, which integrates Pyramid Pathway Input, CoordConv convolution, and Dual-Inut Cross Attention (DCA) modules for optimized performance. Initially, the Pyramid Pathway Input equips the model to identify features at multiple scales, markedly enhancing its ability to discriminate between roads and other background elements. Secondly, by adopting CoordConv convolutional layers, the model achieves heightened accuracy in road recognition and extraction against complex backdrops. Moreover, the DCA module serves dual purposes: it is employed at the encoder stage to efficiently consolidate feature maps across scales, thereby fortifying the model’s road detection capabilities while mitigating false positives. In the skip connection stages, the DCA module further refines the continuity and accuracy of the features. Extensive empirical evaluation substantiates that PCCAU-Net significantly outperforms existing state-of-the-art techniques on multiple benchmarks, including precision, recall, and Intersection-over-Union(IoU). Consequently, PCCAU-Net not only represents a considerable advancement in road extraction research, but also demonstrates vast potential for broader applications, such as urban planning and traffic analytics. |
| first_indexed | 2024-03-07T22:43:31Z |
| format | Article |
| id | doaj.art-67dbf984b2b549eca6697520530fc37b |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-07T22:43:31Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-67dbf984b2b549eca6697520530fc37b2024-02-23T15:06:37ZengMDPI AGApplied Sciences2076-34172024-02-01144163410.3390/app14041634PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA ModuleXiaoqin Xue0Chao Ren1Anchao Yin2Ying Zhou3Yuanyuan Liu4Cong Ding5Jiakai Lu6College of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaCollege of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, ChinaIn the domain of remote sensing research, the extraction of roads from high-resolution imagery remains a formidable challenge. In this paper, we introduce an advanced architecture called PCCAU-Net, which integrates Pyramid Pathway Input, CoordConv convolution, and Dual-Inut Cross Attention (DCA) modules for optimized performance. Initially, the Pyramid Pathway Input equips the model to identify features at multiple scales, markedly enhancing its ability to discriminate between roads and other background elements. Secondly, by adopting CoordConv convolutional layers, the model achieves heightened accuracy in road recognition and extraction against complex backdrops. Moreover, the DCA module serves dual purposes: it is employed at the encoder stage to efficiently consolidate feature maps across scales, thereby fortifying the model’s road detection capabilities while mitigating false positives. In the skip connection stages, the DCA module further refines the continuity and accuracy of the features. Extensive empirical evaluation substantiates that PCCAU-Net significantly outperforms existing state-of-the-art techniques on multiple benchmarks, including precision, recall, and Intersection-over-Union(IoU). Consequently, PCCAU-Net not only represents a considerable advancement in road extraction research, but also demonstrates vast potential for broader applications, such as urban planning and traffic analytics.https://www.mdpi.com/2076-3417/14/4/1634road extractionpyramid pathway inputCoordConvDual-Input Cross Attention (DCA) Modules |
| spellingShingle | Xiaoqin Xue Chao Ren Anchao Yin Ying Zhou Yuanyuan Liu Cong Ding Jiakai Lu PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module Applied Sciences road extraction pyramid pathway input CoordConv Dual-Input Cross Attention (DCA) Modules |
| title | PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module |
| title_full | PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module |
| title_fullStr | PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module |
| title_full_unstemmed | PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module |
| title_short | PCCAU-Net: A Novel Road Extraction Method Based on Coord Convolution and a DCA Module |
| title_sort | pccau net a novel road extraction method based on coord convolution and a dca module |
| topic | road extraction pyramid pathway input CoordConv Dual-Input Cross Attention (DCA) Modules |
| url | https://www.mdpi.com/2076-3417/14/4/1634 |
| work_keys_str_mv | AT xiaoqinxue pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT chaoren pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT anchaoyin pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT yingzhou pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT yuanyuanliu pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT congding pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule AT jiakailu pccaunetanovelroadextractionmethodbasedoncoordconvolutionandadcamodule |