U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil
Currently, there exists a growing demand for individual building mapping in regions of rapid urban growth in less-developed countries. Most existing methods can segment buildings but cannot discriminate adjacent buildings. Here, we present a new convolutional neural network architecture (CNN) called...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/10/1544 |
| _version_ | 1797568177267277824 |
|---|---|
| author | Fabien H. Wagner Ricardo Dalagnol Yuliya Tarabalka Tassiana Y. F. Segantine Rogério Thomé Mayumi C. M. Hirye |
| author_facet | Fabien H. Wagner Ricardo Dalagnol Yuliya Tarabalka Tassiana Y. F. Segantine Rogério Thomé Mayumi C. M. Hirye |
| author_sort | Fabien H. Wagner |
| collection | DOAJ |
| description | Currently, there exists a growing demand for individual building mapping in regions of rapid urban growth in less-developed countries. Most existing methods can segment buildings but cannot discriminate adjacent buildings. Here, we present a new convolutional neural network architecture (CNN) called U-net-id that performs building instance segmentation. The proposed network is trained with WorldView-3 satellite RGB images (0.3 m) and three different labeled masks. The first is the building mask; the second is the border mask, which is the border of the building segment with 4 pixels added outside and 3 pixels inside; and the third is the inner segment mask, which is the segment of the building diminished by 2 pixels. The architecture consists of three parallel paths, one for each mask, all starting with a U-net model. To accurately capture the overlap between the masks, all activation layers of the U-nets are copied and concatenated on each path and sent to two additional convolutional layers before the output activation layers. The method was tested with a dataset of 7563 manually delineated individual buildings of the city of Joanópolis-SP, Brazil. On this dataset, the semantic segmentation showed an overall accuracy of 97.67% and an F1-Score of 0.937 and the building individual instance segmentation showed good performance with a mean intersection over union (IoU) of 0.582 (median IoU = 0.694). |
| first_indexed | 2024-03-10T19:52:45Z |
| format | Article |
| id | doaj.art-949a457aa76b4bc580b801bc9c3a8783 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T19:52:45Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-949a457aa76b4bc580b801bc9c3a87832023-11-20T00:13:59ZengMDPI AGRemote Sensing2072-42922020-05-011210154410.3390/rs12101544U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, BrazilFabien H. Wagner0Ricardo Dalagnol1Yuliya Tarabalka2Tassiana Y. F. Segantine3Rogério Thomé4Mayumi C. M. Hirye5GeoProcessing Division, Foundation for Science, Technology and Space Applications—FUNCATE, São José dos Campos, SP 12210-131, BrazilRemote Sensing Division, National Institute for Space Research—INPE, São José dos Campos, SP 12227-010, BrazilLuxcarta Technology, Parc d’Activité l’Argile, Lot 119b, 06370 Mouans Sartoux, FranceGeoProcessing Division, Foundation for Science, Technology and Space Applications—FUNCATE, São José dos Campos, SP 12210-131, BrazilGeoProcessing Division, Foundation for Science, Technology and Space Applications—FUNCATE, São José dos Campos, SP 12210-131, BrazilQuapá Lab, Faculty of Architecture and Urbanism, University of São Paulo—USP, São Paulo, SP 05508-900, BrazilCurrently, there exists a growing demand for individual building mapping in regions of rapid urban growth in less-developed countries. Most existing methods can segment buildings but cannot discriminate adjacent buildings. Here, we present a new convolutional neural network architecture (CNN) called U-net-id that performs building instance segmentation. The proposed network is trained with WorldView-3 satellite RGB images (0.3 m) and three different labeled masks. The first is the building mask; the second is the border mask, which is the border of the building segment with 4 pixels added outside and 3 pixels inside; and the third is the inner segment mask, which is the segment of the building diminished by 2 pixels. The architecture consists of three parallel paths, one for each mask, all starting with a U-net model. To accurately capture the overlap between the masks, all activation layers of the U-nets are copied and concatenated on each path and sent to two additional convolutional layers before the output activation layers. The method was tested with a dataset of 7563 manually delineated individual buildings of the city of Joanópolis-SP, Brazil. On this dataset, the semantic segmentation showed an overall accuracy of 97.67% and an F1-Score of 0.937 and the building individual instance segmentation showed good performance with a mean intersection over union (IoU) of 0.582 (median IoU = 0.694).https://www.mdpi.com/2072-4292/12/10/1544instance segmentationU-netbuilding detectionurban landscape |
| spellingShingle | Fabien H. Wagner Ricardo Dalagnol Yuliya Tarabalka Tassiana Y. F. Segantine Rogério Thomé Mayumi C. M. Hirye U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil Remote Sensing instance segmentation U-net building detection urban landscape |
| title | U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil |
| title_full | U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil |
| title_fullStr | U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil |
| title_full_unstemmed | U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil |
| title_short | U-Net-Id, an Instance Segmentation Model for Building Extraction from Satellite Images—Case Study in the Joanópolis City, Brazil |
| title_sort | u net id an instance segmentation model for building extraction from satellite images case study in the joanopolis city brazil |
| topic | instance segmentation U-net building detection urban landscape |
| url | https://www.mdpi.com/2072-4292/12/10/1544 |
| work_keys_str_mv | AT fabienhwagner unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil AT ricardodalagnol unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil AT yuliyatarabalka unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil AT tassianayfsegantine unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil AT rogeriothome unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil AT mayumicmhirye unetidaninstancesegmentationmodelforbuildingextractionfromsatelliteimagescasestudyinthejoanopoliscitybrazil |