Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark
Mountain roads are of great significance to traffic navigation and military road planning. Extracting mountain roads based on high-resolution remote sensing images (HRSIs) is a hot spot in current road extraction research. However, massive terrain objects, blurred road edges, and sand coverage in co...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/19/4729 |
| _version_ | 1827653324224069632 |
|---|---|
| author | Xinyu Zhang Yu Jiang Lizhe Wang Wei Han Ruyi Feng Runyu Fan Sheng Wang |
| author_facet | Xinyu Zhang Yu Jiang Lizhe Wang Wei Han Ruyi Feng Runyu Fan Sheng Wang |
| author_sort | Xinyu Zhang |
| collection | DOAJ |
| description | Mountain roads are of great significance to traffic navigation and military road planning. Extracting mountain roads based on high-resolution remote sensing images (HRSIs) is a hot spot in current road extraction research. However, massive terrain objects, blurred road edges, and sand coverage in complex environments make it challenging to extract mountain roads from HRSIs. Complex environments result in weak research results on targeted extraction models and a lack of corresponding datasets. To solve the above problems, first, we propose a new dataset: Road Datasets in Complex Mountain Environments (RDCME). RDCME comes from the QuickBird satellite, which is at an elevation between 1264 m and 1502 m with a resolution of 0.61 m; it contains 775 image samples, including red, green, and blue channels. Then, we propose the Light Roadformer model, which uses a transformer module and self-attention module to focus on extracting more accurate road edge information. A post-process module is further used to remove incorrectly predicted road segments. Compared with previous related models, the Light Roadformer proposed in this study has higher accuracy. Light Roadformer achieved the highest IoU of 89.5% for roads on the validation set and 88.8% for roads on the test set. The test on RDCME using Light Roadformer shows that the results of this study have broad application prospects in the extraction of mountain roads with similar backgrounds. |
| first_indexed | 2024-03-09T21:15:24Z |
| format | Article |
| id | doaj.art-df9ee5f7b6a14d268a1f73a636e36843 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T21:15:24Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-df9ee5f7b6a14d268a1f73a636e368432023-11-23T21:37:38ZengMDPI AGRemote Sensing2072-42922022-09-011419472910.3390/rs14194729Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New BenchmarkXinyu Zhang0Yu Jiang1Lizhe Wang2Wei Han3Ruyi Feng4Runyu Fan5Sheng Wang6School of Computer Science, China University of Geosciences, Wuhan 430078, ChinaInstitute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaMountain roads are of great significance to traffic navigation and military road planning. Extracting mountain roads based on high-resolution remote sensing images (HRSIs) is a hot spot in current road extraction research. However, massive terrain objects, blurred road edges, and sand coverage in complex environments make it challenging to extract mountain roads from HRSIs. Complex environments result in weak research results on targeted extraction models and a lack of corresponding datasets. To solve the above problems, first, we propose a new dataset: Road Datasets in Complex Mountain Environments (RDCME). RDCME comes from the QuickBird satellite, which is at an elevation between 1264 m and 1502 m with a resolution of 0.61 m; it contains 775 image samples, including red, green, and blue channels. Then, we propose the Light Roadformer model, which uses a transformer module and self-attention module to focus on extracting more accurate road edge information. A post-process module is further used to remove incorrectly predicted road segments. Compared with previous related models, the Light Roadformer proposed in this study has higher accuracy. Light Roadformer achieved the highest IoU of 89.5% for roads on the validation set and 88.8% for roads on the test set. The test on RDCME using Light Roadformer shows that the results of this study have broad application prospects in the extraction of mountain roads with similar backgrounds.https://www.mdpi.com/2072-4292/14/19/4729road extractionremote sensinghigh-resolution remote sensingsemantic segmentationtransformer |
| spellingShingle | Xinyu Zhang Yu Jiang Lizhe Wang Wei Han Ruyi Feng Runyu Fan Sheng Wang Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark Remote Sensing road extraction remote sensing high-resolution remote sensing semantic segmentation transformer |
| title | Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark |
| title_full | Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark |
| title_fullStr | Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark |
| title_full_unstemmed | Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark |
| title_short | Complex Mountain Road Extraction in High-Resolution Remote Sensing Images via a Light Roadformer and a New Benchmark |
| title_sort | complex mountain road extraction in high resolution remote sensing images via a light roadformer and a new benchmark |
| topic | road extraction remote sensing high-resolution remote sensing semantic segmentation transformer |
| url | https://www.mdpi.com/2072-4292/14/19/4729 |
| work_keys_str_mv | AT xinyuzhang complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT yujiang complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT lizhewang complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT weihan complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT ruyifeng complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT runyufan complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark AT shengwang complexmountainroadextractioninhighresolutionremotesensingimagesviaalightroadformerandanewbenchmark |