In the high-resolution images, the undamaged buildings generally show a natural textural feature, while the dam- aged or semi-damaged buildings always exhibit some low-grayscale blocks because of their coarsely damag...In the high-resolution images, the undamaged buildings generally show a natural textural feature, while the dam- aged or semi-damaged buildings always exhibit some low-grayscale blocks because of their coarsely damaged sections. If we use a proper threshold to classify the grayscale of image, some independent holes will appear in the damaged regions. By using such statistical information as the number of holes in every region, or the ratio between the area of holes and that of the region, etc, the damaged buildings can be separated from the undamaged, thus automatic detection of damaged buildings can be realized. Based on these characteristics, a new method to auto- matically detect the damage buildings by using regional structure and statistical information of texture is presented in the paper. In order to test its validity, 1-m-resolution iKonos merged image of the 2001 Bhuj earthquake and grayscale aerial photos of the 1976 Tangshan earthquake are selected as two examples to automatically detect the damaged buildings. Satisfied results are obtained.展开更多
Recent studies for computer vision and deep learning-based,post-earthquake inspections on RC structures mainly perform well for specific tasks,while the trained models must be fine-tuned and re-trained when facing new...Recent studies for computer vision and deep learning-based,post-earthquake inspections on RC structures mainly perform well for specific tasks,while the trained models must be fine-tuned and re-trained when facing new tasks and datasets,which is inevitably time-consuming.This study proposes a multi-task learning approach that simultaneously accomplishes the semantic segmentation of seven-type structural components,three-type seismic damage,and four-type deterioration states.The proposed method contains a CNN-based encoder-decoder backbone subnetwork with skip-connection modules and a multi-head,task-specific recognition subnetwork.The backbone subnetwork is designed to extract multi-level features of post-earthquake RC structures.The multi-head,task-specific recognition subnetwork consists of three individual self-attention pipelines,each of which utilizes extracted multi-level features from the backbone network as a mutual guidance for the individual segmentation task.A synthetical loss function is designed with real-time adaptive coefficients to balance multi-task losses and focus on the most unstably fluctuating one.Ablation experiments and comparative studies are further conducted to demonstrate their effectiveness and necessity.The results show that the proposed method can simultaneously recognize different structural components,seismic damage,and deterioration states,and that the overall performance of the three-task learning models gains general improvement when compared to all single-task and dual-task models.展开更多
文摘In the high-resolution images, the undamaged buildings generally show a natural textural feature, while the dam- aged or semi-damaged buildings always exhibit some low-grayscale blocks because of their coarsely damaged sections. If we use a proper threshold to classify the grayscale of image, some independent holes will appear in the damaged regions. By using such statistical information as the number of holes in every region, or the ratio between the area of holes and that of the region, etc, the damaged buildings can be separated from the undamaged, thus automatic detection of damaged buildings can be realized. Based on these characteristics, a new method to auto- matically detect the damage buildings by using regional structure and statistical information of texture is presented in the paper. In order to test its validity, 1-m-resolution iKonos merged image of the 2001 Bhuj earthquake and grayscale aerial photos of the 1976 Tangshan earthquake are selected as two examples to automatically detect the damaged buildings. Satisfied results are obtained.
基金National Key R&D Program of China under Grant No.2019YFC1511005the National Natural Science Foundation of China under Grant Nos.51921006,52192661 and 52008138+2 种基金the China Postdoctoral Science Foundation under Grant Nos.BX20190102 and 2019M661286the Heilongjiang Natural Science Foundation under Grant No.LH2022E070the Heilongjiang Province Postdoctoral Science Foundation under Grant Nos.LBH-TZ2016 and LBH-Z19064。
文摘Recent studies for computer vision and deep learning-based,post-earthquake inspections on RC structures mainly perform well for specific tasks,while the trained models must be fine-tuned and re-trained when facing new tasks and datasets,which is inevitably time-consuming.This study proposes a multi-task learning approach that simultaneously accomplishes the semantic segmentation of seven-type structural components,three-type seismic damage,and four-type deterioration states.The proposed method contains a CNN-based encoder-decoder backbone subnetwork with skip-connection modules and a multi-head,task-specific recognition subnetwork.The backbone subnetwork is designed to extract multi-level features of post-earthquake RC structures.The multi-head,task-specific recognition subnetwork consists of three individual self-attention pipelines,each of which utilizes extracted multi-level features from the backbone network as a mutual guidance for the individual segmentation task.A synthetical loss function is designed with real-time adaptive coefficients to balance multi-task losses and focus on the most unstably fluctuating one.Ablation experiments and comparative studies are further conducted to demonstrate their effectiveness and necessity.The results show that the proposed method can simultaneously recognize different structural components,seismic damage,and deterioration states,and that the overall performance of the three-task learning models gains general improvement when compared to all single-task and dual-task models.
