Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery
Urban research is progressively moving towards fine-grained simulation and requires more granular and accurate geospatial data. In comparison to building footprints, roof structure lines (RSLs) are finer-grained elements of building roofs that provide a more sophisticated data reference. However, ge...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-03-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S030324342200006X |
| _version_ | 1797961344390004736 |
|---|---|
| author | Zhen Qian Min Chen Teng Zhong Fan Zhang Rui Zhu Zhixin Zhang Kai Zhang Zhuo Sun Guonian Lü |
| author_facet | Zhen Qian Min Chen Teng Zhong Fan Zhang Rui Zhu Zhixin Zhang Kai Zhang Zhuo Sun Guonian Lü |
| author_sort | Zhen Qian |
| collection | DOAJ |
| description | Urban research is progressively moving towards fine-grained simulation and requires more granular and accurate geospatial data. In comparison to building footprints, roof structure lines (RSLs) are finer-grained elements of building roofs that provide a more sophisticated data reference. However, generating high-quality and up-to-date RSLs is arduous owing to the high expense of data sources (e.g., digital surface models and light detection and ranging data) and the low robustness of conventional image processing approaches. While the current combination of high-resolution satellite imagery and deep learning methods enables the automatic generation of RSLs, it also introduces two distinct challenges. First, the high diversity of roof sizes, forms, and spatial distribution complicates the extraction of essential RSL features from satellite imagery using general deep learning methods. Second, the significant class imbalance issue between foreground objects (i.e., RSLs) and background context in satellite imagery makes it difficult for deep learning methods to concentrate on RSL locations. To overcome these challenges and effectively delineate RSLs from satellite imagery, this study designs Deep Roof Refiner—an end-to-end and detail-oriented deep learning network and proposes a synthetic strategy to enhance the network’s performance. The effectiveness of the proposed network is verified by quantitative and qualitative experiments, with the optimal dataset scale F1-score and optimal image scale F1-score of 60.89% and 63.48%, respectively. The proposed network significantly outperforms state-of-the-art deep learning methods and associated conventional research. The results indicate that the delineated RSLs can serve as a reliable data source for some urban building-based studies. |
| first_indexed | 2024-04-11T00:57:04Z |
| format | Article |
| id | doaj.art-e0c70be7aec94691a1263db5ce3c308b |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-04-11T00:57:04Z |
| publishDate | 2022-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-e0c70be7aec94691a1263db5ce3c308b2023-01-05T04:31:02ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322022-03-01107102680Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imageryZhen Qian0Min Chen1Teng Zhong2Fan Zhang3Rui Zhu4Zhixin Zhang5Kai Zhang6Zhuo Sun7Guonian Lü8Key Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaKey Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, China; Corresponding author at: School of Geography, Nanjing Normal University, NO.1, Wenyuan Road, Qixia District, Nanjing 210023, China.Key Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaSenseable City Lab, Massachusetts Institute of Technology, Cambridge, MA 02139, USADepartment of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Kowloon, Hong Kong, ChinaCollege of Geography & Marine, Nanjing University, Nanjing, PO Box 2100913, ChinaKey Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaKey Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaKey Laboratory of Virtual Geographic Environment (Ministry of Education of PRC), Nanjing Normal University, Nanjing 210023, China; State Key Laboratory Cultivation Base of Geographical Environment Evolution, Nanjing 210023, China; Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaUrban research is progressively moving towards fine-grained simulation and requires more granular and accurate geospatial data. In comparison to building footprints, roof structure lines (RSLs) are finer-grained elements of building roofs that provide a more sophisticated data reference. However, generating high-quality and up-to-date RSLs is arduous owing to the high expense of data sources (e.g., digital surface models and light detection and ranging data) and the low robustness of conventional image processing approaches. While the current combination of high-resolution satellite imagery and deep learning methods enables the automatic generation of RSLs, it also introduces two distinct challenges. First, the high diversity of roof sizes, forms, and spatial distribution complicates the extraction of essential RSL features from satellite imagery using general deep learning methods. Second, the significant class imbalance issue between foreground objects (i.e., RSLs) and background context in satellite imagery makes it difficult for deep learning methods to concentrate on RSL locations. To overcome these challenges and effectively delineate RSLs from satellite imagery, this study designs Deep Roof Refiner—an end-to-end and detail-oriented deep learning network and proposes a synthetic strategy to enhance the network’s performance. The effectiveness of the proposed network is verified by quantitative and qualitative experiments, with the optimal dataset scale F1-score and optimal image scale F1-score of 60.89% and 63.48%, respectively. The proposed network significantly outperforms state-of-the-art deep learning methods and associated conventional research. The results indicate that the delineated RSLs can serve as a reliable data source for some urban building-based studies.http://www.sciencedirect.com/science/article/pii/S030324342200006XDeep learningRoof Structure LinesSatellite ImageryFine-grained Geospatial Data |
| spellingShingle | Zhen Qian Min Chen Teng Zhong Fan Zhang Rui Zhu Zhixin Zhang Kai Zhang Zhuo Sun Guonian Lü Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery International Journal of Applied Earth Observations and Geoinformation Deep learning Roof Structure Lines Satellite Imagery Fine-grained Geospatial Data |
| title | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| title_full | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| title_fullStr | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| title_full_unstemmed | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| title_short | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| title_sort | deep roof refiner a detail oriented deep learning network for refined delineation of roof structure lines using satellite imagery |
| topic | Deep learning Roof Structure Lines Satellite Imagery Fine-grained Geospatial Data |
| url | http://www.sciencedirect.com/science/article/pii/S030324342200006X |
| work_keys_str_mv | AT zhenqian deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT minchen deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT tengzhong deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT fanzhang deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT ruizhu deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT zhixinzhang deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT kaizhang deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT zhuosun deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery AT guonianlu deeproofrefineradetailorienteddeeplearningnetworkforrefineddelineationofroofstructurelinesusingsatelliteimagery |