@Article{WagnerSiTaSeThHi:2020:UnInSe,
               author = "Wagner, Fabien Hubert and Silva, Ricardo Dal Agnol da and 
                         Tarabalka, Yuliya and Segantine, Tassiana Y. F. and Thom{\'e}, 
                         Rog{\'e}rio and Hirye, Mayumi C. M.",
          affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto 
                         Nacional de Pesquisas Espaciais (INPE)} and {Luxcarta Technology} 
                         and Funda{\c{c}}{\~a}o de Ci{\^e}ncia, Aplica{\c{c}}{\~o}es e 
                         Tecnologia Espaciais (FUNCATE) and Funda{\c{c}}{\~a}o de 
                         Ci{\^e}ncia, Aplica{\c{c}}{\~o}es e Tecnologia Espaciais 
                         (FUNCATE) and {Universidade de S{\~a}o Paulo (USP)}",
                title = "U-net-id, an instance segmentation model for building extraction 
                         from satellite images-Case study in the Joanopolis City, Brazil",
              journal = "Remote Sensing",
                 year = "2020",
               volume = "12",
               number = "10",
                pages = "e1544",
                month = "May",
             keywords = "instance segmentation, U-net, building detection, urban 
                         landscape.",
             abstract = "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 
                         Joan{\'o}polis-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).",
                  doi = "10.3390/rs12101544",
                  url = "http://dx.doi.org/10.3390/rs12101544",
                 issn = "2072-4292",
             language = "en",
           targetfile = "remotesensing-12-01544-v2.pdf",
        urlaccessdate = "11 maio 2024"
}