Large-scale optical sensing and precise, rapid assessment of seismic building damage in urban communities are increasingly demanded in disaster prevention and reduction. The common method is to train a convolutional neural network (CNN) in a pixel-level semantic segmentation approach and does not fully consider the characteristics of the assessment objectives. This study developed a machine-learning-derived two-stage method for post-earthquake building location and damage assessment considering the data characteristics of satellite remote sensing (SRS) optical images with dense distribution, small size, and imbalanced numbers. It included a modified You Only Look Once (YOLOv4) object detection module and a support vector machine (SVM) based classification module. In the primary step, the multiscale features were successfully extracted and fused from SRS images of densely distributed buildings by optimizing the YOLOv4 model toward the network structures, training hyperparameters, and anchor boxes. The fusion improved multi-channel features, optimization of network structure and hyperparameters have significantly enhanced the average location accuracy of post-earthquake buildings. Thereafter, three statistics (i.e., the angular second moment, dissimilarity, and inverse difference moment) were further discovered to effectively extract the characteristic value for earthquake damage from located buildings in SRS optical images based on the gray level co-occurrence matrix. They were used as the texture features to distinguish damage intensities of buildings, using the SVM model. The investigated dataset included 386 pre- and post-earthquake SRS optical images of the 2017 Mexico City earthquake, with a resolution of 1024 × 1024 pixels. Results show that the average location accuracy of post-earthquake buildings exceeds 95.7% and that the binary classification accuracy for damage assessment reaches 97.1%. The proposed two-stage method was validated by its extremely high precision in respect of densely distributed small buildings, indicating the promising potential of computer vision in large-scale disaster prevention and reduction using SRS datasets.

Timely acquiring accurate information on the earthquake-induced damage of buildings is crucial for emergency assessment and post-disaster rescue. Optical remote sensing photography has been a typical method for obtaining seismic data in the early stage after an earthquake due to its wide coverage and fast response speed. Currently, convolutional neural networks (CNNs) are widely applied for remote sensing image recognition. However, insufficient extraction and expression ability of global correlations between local image patches limit the performance of dense building segmentation under real-world scenarios. To address the above issue, this paper proposes an improved Swin Transformer to segment dense urban buildings from remote sensing images with complex backgrounds. The original Swin Transformer is used as a backbone of the encoder, and a convolutional block attention module (CBAM) is employed in the linear embedding and patch merging stages to focus on significant features. Hierarchical feature maps are first fused by convolution operators to strengthen the feature extraction process and adjust the channel of feature dimension, followed by being fed to the UPerNet (as the decoder) to obtain the final segmentation map. Buildings are labeled from the collected remote sensing images of the Yushu and Beichuan earthquakes and classified into collapsed and non-collapsed buildings. Data augmentation transformations of horizontal and vertical flipping, brightness adjustment, uniform fogging, and non-uniform fogging are performed to simulate different photography directions, light overexposure, darkness, and foggy occlusions under actual situations. The effectiveness and superiority of the proposed method over the original Swin Transformer and several mature CNN-based segmentation models are validated by ablation experiments and comparative studies. The results show that the mean intersection over union (mIoU) of the improved Swin Transformer reaches 88.53%, achieving a significant improvement of 1.3% compared to the original Swin Transformer of 87.23%. Furthermore, the improved Swin Transformer can accurately identify dense buildings in remote sensing images under complex weather disturbances. Finally, the results indicate good stability, robustness, and generalization ability of the proposed method in the dense segmentation of seismic buildings.

No abstract

Deep‐learning‐based automatic recognition of post‐earthquake damage for urban buildings is increasingly in demand for rapid and precise assessment of seismic hazards from optical remote sensing images. In this study, a novel loss function fusing geometric consistency constraint (GCC) with cross‐entropy (CE) loss is designed for post‐earthquake building segmentation with complex geometric features across multiple scales. Specifically, the GCC loss incorporates three critical components, namely, split line length, curvature, and area, and enables the exact extraction of the geometric constraints of boundary and region for damaged buildings. Through the optimization of multiple key coefficients of GCC loss, the proposed method achieves significant performance improvements in semantic segmentation, which is attributed to the enhanced ability to identify and capture the pixel relationship near the boundary. Merging GCC in the loss function enables faster and more accurate convergence of predicted values towards the ground truth during the training process, surpassing the performance of the CE loss alone. The results show that the combination of GCC and CE losses achieves the largest validation mIoU of 86.98% for damaged buildings segmentation, which facilitates post‐earthquake assessment with high accuracy. Moreover, incorporating GCC leads to more precise and robust seismic damage segmentation by effectively improving edge closure, removing internal noise, and reducing false‐positive and false‐negative misrecognition. In addition, the GCC term further validates the effectiveness of improving segmentation tasks for other networks (e.g., DeepLabv3+). The GCC‐derived method exhibits its desirable performance on segmentation accuracy, portability, and universality for building recognition with complex geometric features and post‐earthquake scenes.

No abstract