Building Extraction from Satellite Images Using Mask R-CNN with Building Boundary Regularization

The DeepGlobe Building Extraction Challenge poses the problem of localizing all building polygons in the given satellite images. We can create polygons using an existing instance segmentation algorithm based on Mask R-CNN. However, polygons produced from instance segmentation have irregular shapes, which are far different from real building footprint boundaries and therefore cannot be directly applied to many cartographic and engineering applications. Hence, we present a method combining Mask R-CNN with building boundary regularization. Through the experiments, we find that the proposed method and Mask R-CNN achieve almost equivalent performance in terms of accuracy and completeness. However, compared to Mask R-CNN, our method produces better regularized polygons which are beneficial in many applications

[1]  Yoonseok Jwa,et al.  AN IMPLICIT REGULARIZATION FOR 3 D BUILDING ROOFTOP MODELING USING AIRBORNE LIDAR DATA , 2012 .

[2]  Bertrand Le Saux,et al.  Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[3]  Pierre Alliez,et al.  High-Resolution Aerial Image Labeling With Convolutional Neural Networks , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[4]  David H. Douglas,et al.  ALGORITHMS FOR THE REDUCTION OF THE NUMBER OF POINTS REQUIRED TO REPRESENT A DIGITIZED LINE OR ITS CARICATURE , 1973 .

[5]  Pierre Alliez,et al.  Polygonization of remote sensing classification maps by mesh approximation , 2017, 2017 IEEE International Conference on Image Processing (ICIP).

[6]  Jaewook Jung,et al.  Implicit Regularization for Reconstructing 3D Building Rooftop Models Using Airborne LiDAR Data , 2017, Sensors.

[7]  Carsten Rother,et al.  InstanceCut: From Edges to Instances with MultiCut , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Geoffrey E. Hinton,et al.  Machine Learning for Aerial Image Labeling , 2013 .

[9]  Yi Li,et al.  Fully Convolutional Instance-Aware Semantic Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Uwe Stilla,et al.  Classification With an Edge: Improving Semantic Image Segmentation with Boundary Detection , 2016, ISPRS Journal of Photogrammetry and Remote Sensing.

[11]  Ronan Collobert,et al.  Learning to Refine Object Segments , 2016, ECCV.

[12]  Kaiming He,et al.  Mask R-CNN , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[13]  Andreas Dengel,et al.  Multi-Task Learning for Segmentation of Building Footprints with Deep Neural Networks , 2017, 2019 IEEE International Conference on Image Processing (ICIP).

[14]  Yoonseok Jwa,et al.  AN IMPLICIT REGULARIZATION FOR 3D BUILDING ROOFTOP MODELING USING AIRBORNE LIDAR DATA , 2012 .

[15]  Yuliya Tarabalka,et al.  End-to-End Learning of Polygons for Remote Sensing Image Classification , 2018, IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium.