Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models
This study focuses on the identification of collapsed buildings in satellite images after earthquakes through deep learning-based image segmentation models. The performance of four different architectures, namely U-Net, LinkNet, FPN, and PSPNet, was evaluated using various performance metrics, such...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/14/3/582 |
| _version_ | 1797241814617423872 |
|---|---|
| author | Kemal Hacıefendioğlu Hasan Basri Başağa Volkan Kahya Korhan Özgan Ahmet Can Altunışık |
| author_facet | Kemal Hacıefendioğlu Hasan Basri Başağa Volkan Kahya Korhan Özgan Ahmet Can Altunışık |
| author_sort | Kemal Hacıefendioğlu |
| collection | DOAJ |
| description | This study focuses on the identification of collapsed buildings in satellite images after earthquakes through deep learning-based image segmentation models. The performance of four different architectures, namely U-Net, LinkNet, FPN, and PSPNet, was evaluated using various performance metrics, such as accuracy, precision, recall, F1 score, specificity, AUC, and IoU. The study used satellite images taken from the area located in the south and southeast of Türkiye covering the eleven provinces which are most affected by the Mw 7.7 Pazarcık (Kahramanmaraş) and Mw 7.6 Elbistan (Kahramanmaraş) earthquakes. The results indicated that FPN and U-Net were the best-performing models depending on the performance metric of interest. FPN achieved the highest accuracy and specificity scores, as well as the best precision score, while U-Net achieved the best recall and F1 score values, as well as the best AUC and IoU scores. The training and validation accuracy and loss curves were analyzed, and the results indicated that all four models achieved an accuracy value of over 96%. The FPN model outperformed the others in terms of accurately segmenting images while maintaining a low loss value. This study provides insights into the potential of deep learning-based image segmentation models in disaster management and can be useful for future research in this field. |
| first_indexed | 2024-04-24T18:29:18Z |
| format | Article |
| id | doaj.art-a490009f7ca84a97a1d601f874becefa |
| institution | Directory Open Access Journal |
| issn | 2075-5309 |
| language | English |
| last_indexed | 2024-04-24T18:29:18Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Buildings |
| spelling | doaj.art-a490009f7ca84a97a1d601f874becefa2024-03-27T13:28:58ZengMDPI AGBuildings2075-53092024-02-0114358210.3390/buildings14030582Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation ModelsKemal Hacıefendioğlu0Hasan Basri Başağa1Volkan Kahya2Korhan Özgan3Ahmet Can Altunışık4Department of Civil Engineering, Karadeniz Technical University, 61080 Trabzon, TurkeyDepartment of Civil Engineering, Karadeniz Technical University, 61080 Trabzon, TurkeyDepartment of Civil Engineering, Karadeniz Technical University, 61080 Trabzon, TurkeyDepartment of Civil Engineering, Karadeniz Technical University, 61080 Trabzon, TurkeyDepartment of Civil Engineering, Karadeniz Technical University, 61080 Trabzon, TurkeyThis study focuses on the identification of collapsed buildings in satellite images after earthquakes through deep learning-based image segmentation models. The performance of four different architectures, namely U-Net, LinkNet, FPN, and PSPNet, was evaluated using various performance metrics, such as accuracy, precision, recall, F1 score, specificity, AUC, and IoU. The study used satellite images taken from the area located in the south and southeast of Türkiye covering the eleven provinces which are most affected by the Mw 7.7 Pazarcık (Kahramanmaraş) and Mw 7.6 Elbistan (Kahramanmaraş) earthquakes. The results indicated that FPN and U-Net were the best-performing models depending on the performance metric of interest. FPN achieved the highest accuracy and specificity scores, as well as the best precision score, while U-Net achieved the best recall and F1 score values, as well as the best AUC and IoU scores. The training and validation accuracy and loss curves were analyzed, and the results indicated that all four models achieved an accuracy value of over 96%. The FPN model outperformed the others in terms of accurately segmenting images while maintaining a low loss value. This study provides insights into the potential of deep learning-based image segmentation models in disaster management and can be useful for future research in this field.https://www.mdpi.com/2075-5309/14/3/582earthquake damage detectioncollapsed building identificationdeep learningimage segmentationsatellite images |
| spellingShingle | Kemal Hacıefendioğlu Hasan Basri Başağa Volkan Kahya Korhan Özgan Ahmet Can Altunışık Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models Buildings earthquake damage detection collapsed building identification deep learning image segmentation satellite images |
| title | Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models |
| title_full | Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models |
| title_fullStr | Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models |
| title_full_unstemmed | Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models |
| title_short | Automatic Detection of Collapsed Buildings after the 6 February 2023 Türkiye Earthquakes Using Post-Disaster Satellite Images with Deep Learning-Based Semantic Segmentation Models |
| title_sort | automatic detection of collapsed buildings after the 6 february 2023 turkiye earthquakes using post disaster satellite images with deep learning based semantic segmentation models |
| topic | earthquake damage detection collapsed building identification deep learning image segmentation satellite images |
| url | https://www.mdpi.com/2075-5309/14/3/582 |
| work_keys_str_mv | AT kemalhacıefendioglu automaticdetectionofcollapsedbuildingsafterthe6february2023turkiyeearthquakesusingpostdisastersatelliteimageswithdeeplearningbasedsemanticsegmentationmodels AT hasanbasribasaga automaticdetectionofcollapsedbuildingsafterthe6february2023turkiyeearthquakesusingpostdisastersatelliteimageswithdeeplearningbasedsemanticsegmentationmodels AT volkankahya automaticdetectionofcollapsedbuildingsafterthe6february2023turkiyeearthquakesusingpostdisastersatelliteimageswithdeeplearningbasedsemanticsegmentationmodels AT korhanozgan automaticdetectionofcollapsedbuildingsafterthe6february2023turkiyeearthquakesusingpostdisastersatelliteimageswithdeeplearningbasedsemanticsegmentationmodels AT ahmetcanaltunısık automaticdetectionofcollapsedbuildingsafterthe6february2023turkiyeearthquakesusingpostdisastersatelliteimageswithdeeplearningbasedsemanticsegmentationmodels |