Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets
Most mainstream research on assessing building damage using satellite imagery is based on scattered datasets and lacks unified standards and methods to quantify and compare the performance of different models. To mitigate these problems, the present study develops a novel end-to-end benchmark model,...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/24/4055 |
| _version_ | 1797544947246694400 |
|---|---|
| author | Yanbing Bai Junjie Hu Jinhua Su Xing Liu Haoyu Liu Xianwen He Shengwang Meng Erick Mas Shunichi Koshimura |
| author_facet | Yanbing Bai Junjie Hu Jinhua Su Xing Liu Haoyu Liu Xianwen He Shengwang Meng Erick Mas Shunichi Koshimura |
| author_sort | Yanbing Bai |
| collection | DOAJ |
| description | Most mainstream research on assessing building damage using satellite imagery is based on scattered datasets and lacks unified standards and methods to quantify and compare the performance of different models. To mitigate these problems, the present study develops a novel end-to-end benchmark model, termed the pyramid pooling module semi-Siamese network (PPM-SSNet), based on a large-scale xBD satellite imagery dataset. The high precision of the proposed model is achieved by adding residual blocks with dilated convolution and squeeze-and-excitation blocks into the network. Simultaneously, the highly automated process of satellite imagery input and damage classification result output is reached by employing concurrent learned attention mechanisms through a semi-Siamese network for end-to-end input and output purposes. Our proposed method achieves F1 scores of 0.90, 0.41, 0.65, and 0.70 for the undamaged, minor-damaged, major-damaged, and destroyed building classes, respectively. From the perspective of end-to-end methods, the ablation experiments and comparative analysis confirm the effectiveness and originality of the PPM-SSNet method. Finally, the consistent prediction results of our model for data from the 2011 Tohoku Earthquake verify the high performance of our model in terms of the domain shift problem, which implies that it is effective for evaluating future disasters. |
| first_indexed | 2024-03-10T14:08:42Z |
| format | Article |
| id | doaj.art-479ef077c76f49cf844a220b1fc5e5c3 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T14:08:42Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-479ef077c76f49cf844a220b1fc5e5c32023-11-21T00:22:37ZengMDPI AGRemote Sensing2072-42922020-12-011224405510.3390/rs12244055Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery DatasetsYanbing Bai0Junjie Hu1Jinhua Su2Xing Liu3Haoyu Liu4Xianwen He5Shengwang Meng6Erick Mas7Shunichi Koshimura8Center for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaShenzhen Institute of Artificial Intelligence and Robotics for Society, Shenzhen 518172, ChinaCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaGraduate School of Information Sciences, Tohoku University, Sendai 980-8579, JapanCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaInternational Research Institute of Disaster Science, Tohoku University, Sendai 980-8579, JapanInternational Research Institute of Disaster Science, Tohoku University, Sendai 980-8579, JapanMost mainstream research on assessing building damage using satellite imagery is based on scattered datasets and lacks unified standards and methods to quantify and compare the performance of different models. To mitigate these problems, the present study develops a novel end-to-end benchmark model, termed the pyramid pooling module semi-Siamese network (PPM-SSNet), based on a large-scale xBD satellite imagery dataset. The high precision of the proposed model is achieved by adding residual blocks with dilated convolution and squeeze-and-excitation blocks into the network. Simultaneously, the highly automated process of satellite imagery input and damage classification result output is reached by employing concurrent learned attention mechanisms through a semi-Siamese network for end-to-end input and output purposes. Our proposed method achieves F1 scores of 0.90, 0.41, 0.65, and 0.70 for the undamaged, minor-damaged, major-damaged, and destroyed building classes, respectively. From the perspective of end-to-end methods, the ablation experiments and comparative analysis confirm the effectiveness and originality of the PPM-SSNet method. Finally, the consistent prediction results of our model for data from the 2011 Tohoku Earthquake verify the high performance of our model in terms of the domain shift problem, which implies that it is effective for evaluating future disasters.https://www.mdpi.com/2072-4292/12/24/4055pyramid pooling modulesemi-Siamesebenchmark modeldamage assessmentend-to-endxBD dataset |
| spellingShingle | Yanbing Bai Junjie Hu Jinhua Su Xing Liu Haoyu Liu Xianwen He Shengwang Meng Erick Mas Shunichi Koshimura Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets Remote Sensing pyramid pooling module semi-Siamese benchmark model damage assessment end-to-end xBD dataset |
| title | Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets |
| title_full | Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets |
| title_fullStr | Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets |
| title_full_unstemmed | Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets |
| title_short | Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets |
| title_sort | pyramid pooling module based semi siamese network a benchmark model for assessing building damage from xbd satellite imagery datasets |
| topic | pyramid pooling module semi-Siamese benchmark model damage assessment end-to-end xBD dataset |
| url | https://www.mdpi.com/2072-4292/12/24/4055 |
| work_keys_str_mv | AT yanbingbai pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT junjiehu pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT jinhuasu pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT xingliu pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT haoyuliu pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT xianwenhe pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT shengwangmeng pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT erickmas pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets AT shunichikoshimura pyramidpoolingmodulebasedsemisiamesenetworkabenchmarkmodelforassessingbuildingdamagefromxbdsatelliteimagerydatasets |