A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images
Deep-learning technologies, especially convolutional neural networks (CNNs), have achieved great success in building extraction from areal images. However, shape details are often lost during the down-sampling process, which results in discontinuous segmentation or inaccurate segmentation boundary....
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/14/2656 |
| _version_ | 1827686966337994752 |
|---|---|
| author | Furong Shi Tong Zhang |
| author_facet | Furong Shi Tong Zhang |
| author_sort | Furong Shi |
| collection | DOAJ |
| description | Deep-learning technologies, especially convolutional neural networks (CNNs), have achieved great success in building extraction from areal images. However, shape details are often lost during the down-sampling process, which results in discontinuous segmentation or inaccurate segmentation boundary. In order to compensate for the loss of shape information, two shape-related auxiliary tasks (i.e., boundary prediction and distance estimation) were jointly learned with building segmentation task in our proposed network. Meanwhile, two consistency constraint losses were designed based on the multi-task network to exploit the duality between the mask prediction and two shape-related information predictions. Specifically, an atrous spatial pyramid pooling (ASPP) module was appended to the top of the encoder of a U-shaped network to obtain multi-scale features. Based on the multi-scale features, one regression loss and two classification losses were used for predicting the distance-transform map, segmentation, and boundary. Two inter-task consistency-loss functions were constructed to ensure the consistency between distance maps and masks, and the consistency between masks and boundary maps. Experimental results on three public aerial image data sets showed that our method achieved superior performance over the recent state-of-the-art models. |
| first_indexed | 2024-03-10T09:26:04Z |
| format | Article |
| id | doaj.art-d522950cbd4b4abf9be17827639e56b3 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T09:26:04Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-d522950cbd4b4abf9be17827639e56b32023-11-22T04:50:26ZengMDPI AGRemote Sensing2072-42922021-07-011314265610.3390/rs13142656A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial ImagesFurong Shi0Tong Zhang1State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaDeep-learning technologies, especially convolutional neural networks (CNNs), have achieved great success in building extraction from areal images. However, shape details are often lost during the down-sampling process, which results in discontinuous segmentation or inaccurate segmentation boundary. In order to compensate for the loss of shape information, two shape-related auxiliary tasks (i.e., boundary prediction and distance estimation) were jointly learned with building segmentation task in our proposed network. Meanwhile, two consistency constraint losses were designed based on the multi-task network to exploit the duality between the mask prediction and two shape-related information predictions. Specifically, an atrous spatial pyramid pooling (ASPP) module was appended to the top of the encoder of a U-shaped network to obtain multi-scale features. Based on the multi-scale features, one regression loss and two classification losses were used for predicting the distance-transform map, segmentation, and boundary. Two inter-task consistency-loss functions were constructed to ensure the consistency between distance maps and masks, and the consistency between masks and boundary maps. Experimental results on three public aerial image data sets showed that our method achieved superior performance over the recent state-of-the-art models.https://www.mdpi.com/2072-4292/13/14/2656building extractionconvolutional neural networkmulti-task learningconsistency constraintsmulti-scale features |
| spellingShingle | Furong Shi Tong Zhang A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images Remote Sensing building extraction convolutional neural network multi-task learning consistency constraints multi-scale features |
| title | A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images |
| title_full | A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images |
| title_fullStr | A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images |
| title_full_unstemmed | A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images |
| title_short | A Multi-Task Network with Distance–Mask–Boundary Consistency Constraints for Building Extraction from Aerial Images |
| title_sort | multi task network with distance mask boundary consistency constraints for building extraction from aerial images |
| topic | building extraction convolutional neural network multi-task learning consistency constraints multi-scale features |
| url | https://www.mdpi.com/2072-4292/13/14/2656 |
| work_keys_str_mv | AT furongshi amultitasknetworkwithdistancemaskboundaryconsistencyconstraintsforbuildingextractionfromaerialimages AT tongzhang amultitasknetworkwithdistancemaskboundaryconsistencyconstraintsforbuildingextractionfromaerialimages AT furongshi multitasknetworkwithdistancemaskboundaryconsistencyconstraintsforbuildingextractionfromaerialimages AT tongzhang multitasknetworkwithdistancemaskboundaryconsistencyconstraintsforbuildingextractionfromaerialimages |