Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow
Deep learning techniques such as convolutional neural networks have largely improved the performance of building segmentation from remote sensing images. However, the images for building segmentation are often in the form of traditional orthophotos, where the relief displacement would cause non-negl...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/1/207 |
| _version_ | 1797497591932387328 |
|---|---|
| author | Qi Chen Yuanyi Zhang Xinyuan Li Pengjie Tao |
| author_facet | Qi Chen Yuanyi Zhang Xinyuan Li Pengjie Tao |
| author_sort | Qi Chen |
| collection | DOAJ |
| description | Deep learning techniques such as convolutional neural networks have largely improved the performance of building segmentation from remote sensing images. However, the images for building segmentation are often in the form of traditional orthophotos, where the relief displacement would cause non-negligible misalignment between the roof outline and the footprint of a building; such misalignment poses considerable challenges for extracting accurate building footprints, especially for high-rise buildings. Aiming at alleviating this problem, a new workflow is proposed for generating rectified building footprints from traditional orthophotos. We first use the facade labels, which are prepared efficiently at low cost, along with the roof labels to train a semantic segmentation network. Then, the well-trained network, which employs the state-of-the-art version of EfficientNet as backbone, extracts the roof segments and the facade segments of buildings from the input image. Finally, after clustering the classified pixels into instance-level building objects and tracing out the roof outlines, an energy function is proposed to drive the roof outline to maximally align with the building footprint; thus, the rectified footprints can be generated. The experiments on the aerial orthophotos covering a high-density residential area in Shanghai demonstrate that the proposed workflow can generate obviously more accurate building footprints than the baseline methods, especially for high-rise buildings. |
| first_indexed | 2024-03-10T03:21:26Z |
| format | Article |
| id | doaj.art-b518d6e2eeab4e3f984add0f06e7ecc3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T03:21:26Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-b518d6e2eeab4e3f984add0f06e7ecc32023-11-23T12:18:43ZengMDPI AGSensors1424-82202021-12-0122120710.3390/s22010207Extracting Rectified Building Footprints from Traditional Orthophotos: A New WorkflowQi Chen0Yuanyi Zhang1Xinyuan Li2Pengjie Tao3School of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, ChinaSchool of Geography and Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430072, ChinaDeep learning techniques such as convolutional neural networks have largely improved the performance of building segmentation from remote sensing images. However, the images for building segmentation are often in the form of traditional orthophotos, where the relief displacement would cause non-negligible misalignment between the roof outline and the footprint of a building; such misalignment poses considerable challenges for extracting accurate building footprints, especially for high-rise buildings. Aiming at alleviating this problem, a new workflow is proposed for generating rectified building footprints from traditional orthophotos. We first use the facade labels, which are prepared efficiently at low cost, along with the roof labels to train a semantic segmentation network. Then, the well-trained network, which employs the state-of-the-art version of EfficientNet as backbone, extracts the roof segments and the facade segments of buildings from the input image. Finally, after clustering the classified pixels into instance-level building objects and tracing out the roof outlines, an energy function is proposed to drive the roof outline to maximally align with the building footprint; thus, the rectified footprints can be generated. The experiments on the aerial orthophotos covering a high-density residential area in Shanghai demonstrate that the proposed workflow can generate obviously more accurate building footprints than the baseline methods, especially for high-rise buildings.https://www.mdpi.com/1424-8220/22/1/207image segmentationbuilding footprintaerial orthophotorelief displacement |
| spellingShingle | Qi Chen Yuanyi Zhang Xinyuan Li Pengjie Tao Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow Sensors image segmentation building footprint aerial orthophoto relief displacement |
| title | Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow |
| title_full | Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow |
| title_fullStr | Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow |
| title_full_unstemmed | Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow |
| title_short | Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow |
| title_sort | extracting rectified building footprints from traditional orthophotos a new workflow |
| topic | image segmentation building footprint aerial orthophoto relief displacement |
| url | https://www.mdpi.com/1424-8220/22/1/207 |
| work_keys_str_mv | AT qichen extractingrectifiedbuildingfootprintsfromtraditionalorthophotosanewworkflow AT yuanyizhang extractingrectifiedbuildingfootprintsfromtraditionalorthophotosanewworkflow AT xinyuanli extractingrectifiedbuildingfootprintsfromtraditionalorthophotosanewworkflow AT pengjietao extractingrectifiedbuildingfootprintsfromtraditionalorthophotosanewworkflow |