Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector
Using aerial cameras, satellite remote sensing or unmanned aerial vehicles (UAV) equipped with cameras can facilitate search and rescue tasks after disasters. The traditional manual interpretation of huge aerial images is inefficient and could be replaced by machine learning-based methods combined w...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2076-3417/9/6/1128 |
| _version_ | 1818536391223541760 |
|---|---|
| author | Yundong Li Wei Hu Han Dong Xueyan Zhang |
| author_facet | Yundong Li Wei Hu Han Dong Xueyan Zhang |
| author_sort | Yundong Li |
| collection | DOAJ |
| description | Using aerial cameras, satellite remote sensing or unmanned aerial vehicles (UAV) equipped with cameras can facilitate search and rescue tasks after disasters. The traditional manual interpretation of huge aerial images is inefficient and could be replaced by machine learning-based methods combined with image processing techniques. Given the development of machine learning, researchers find that convolutional neural networks can effectively extract features from images. Some target detection methods based on deep learning, such as the single-shot multibox detector (SSD) algorithm, can achieve better results than traditional methods. However, the impressive performance of machine learning-based methods results from the numerous labeled samples. Given the complexity of post-disaster scenarios, obtaining many samples in the aftermath of disasters is difficult. To address this issue, a damaged building assessment method using SSD with pretraining and data augmentation is proposed in the current study and highlights the following aspects. (1) Objects can be detected and classified into undamaged buildings, damaged buildings, and ruins. (2) A convolution auto-encoder (CAE) that consists of VGG16 is constructed and trained using unlabeled post-disaster images. As a transfer learning strategy, the weights of the SSD model are initialized using the weights of the CAE counterpart. (3) Data augmentation strategies, such as image mirroring, rotation, Gaussian blur, and Gaussian noise processing, are utilized to augment the training data set. As a case study, aerial images of Hurricane Sandy in 2012 were maximized to validate the proposed method’s effectiveness. Experiments show that the pretraining strategy can improve of 10% in terms of overall accuracy compared with the SSD trained from scratch. These experiments also demonstrate that using data augmentation strategies can improve mAP and mF1 by 72% and 20%, respectively. Finally, the experiment is further verified by another dataset of Hurricane Irma, and it is concluded that the paper method is feasible. |
| first_indexed | 2024-12-11T18:37:21Z |
| format | Article |
| id | doaj.art-8d0488eb44c8403caac0034fb3a3e8f5 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-12-11T18:37:21Z |
| publishDate | 2019-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-8d0488eb44c8403caac0034fb3a3e8f52022-12-22T00:54:43ZengMDPI AGApplied Sciences2076-34172019-03-0196112810.3390/app9061128app9061128Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox DetectorYundong Li0Wei Hu1Han Dong2Xueyan Zhang3School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaUsing aerial cameras, satellite remote sensing or unmanned aerial vehicles (UAV) equipped with cameras can facilitate search and rescue tasks after disasters. The traditional manual interpretation of huge aerial images is inefficient and could be replaced by machine learning-based methods combined with image processing techniques. Given the development of machine learning, researchers find that convolutional neural networks can effectively extract features from images. Some target detection methods based on deep learning, such as the single-shot multibox detector (SSD) algorithm, can achieve better results than traditional methods. However, the impressive performance of machine learning-based methods results from the numerous labeled samples. Given the complexity of post-disaster scenarios, obtaining many samples in the aftermath of disasters is difficult. To address this issue, a damaged building assessment method using SSD with pretraining and data augmentation is proposed in the current study and highlights the following aspects. (1) Objects can be detected and classified into undamaged buildings, damaged buildings, and ruins. (2) A convolution auto-encoder (CAE) that consists of VGG16 is constructed and trained using unlabeled post-disaster images. As a transfer learning strategy, the weights of the SSD model are initialized using the weights of the CAE counterpart. (3) Data augmentation strategies, such as image mirroring, rotation, Gaussian blur, and Gaussian noise processing, are utilized to augment the training data set. As a case study, aerial images of Hurricane Sandy in 2012 were maximized to validate the proposed method’s effectiveness. Experiments show that the pretraining strategy can improve of 10% in terms of overall accuracy compared with the SSD trained from scratch. These experiments also demonstrate that using data augmentation strategies can improve mAP and mF1 by 72% and 20%, respectively. Finally, the experiment is further verified by another dataset of Hurricane Irma, and it is concluded that the paper method is feasible.http://www.mdpi.com/2076-3417/9/6/1128building damage assessmentpost-eventdeep learningSSDconvolutional autoencoder |
| spellingShingle | Yundong Li Wei Hu Han Dong Xueyan Zhang Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector Applied Sciences building damage assessment post-event deep learning SSD convolutional autoencoder |
| title | Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector |
| title_full | Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector |
| title_fullStr | Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector |
| title_full_unstemmed | Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector |
| title_short | Building Damage Detection from Post-Event Aerial Imagery Using Single Shot Multibox Detector |
| title_sort | building damage detection from post event aerial imagery using single shot multibox detector |
| topic | building damage assessment post-event deep learning SSD convolutional autoencoder |
| url | http://www.mdpi.com/2076-3417/9/6/1128 |
| work_keys_str_mv | AT yundongli buildingdamagedetectionfromposteventaerialimageryusingsingleshotmultiboxdetector AT weihu buildingdamagedetectionfromposteventaerialimageryusingsingleshotmultiboxdetector AT handong buildingdamagedetectionfromposteventaerialimageryusingsingleshotmultiboxdetector AT xueyanzhang buildingdamagedetectionfromposteventaerialimageryusingsingleshotmultiboxdetector |