Towards a rapid automatic detection of building damage using remote sensing for disaster management
Bibliographic record
Abstract
Purpose – Satellite and airborne images are increasingly used at different stages of disaster management, especially in the detection of infrastructure damage. Although semi- or full automatic techniques to detect damage have been proposed, they have not been used in emergency situations. Damage maps produced by international organisations are still based on visual interpretation of images, which is time- and labour-consuming. The purpose of this paper is to investigate how an automatic mapping of damage can be helpful for a first and rapid assessment of building damage. Design/methodology/approach – The study area is located in Port-au-Prince (Haiti) stricken by an earthquake in January 2010. To detect building damage, the paper uses optical images (15 cm of spatial resolution) coupled with height data (LiDAR, 1 m of spatial resolution). By undertaking an automatic object-oriented classification, the paper identifies three categories of building damages: intact buildings, collapsed buildings and debris. Findings – Data processing for the study area covering 11 km2 took about 15 hours. The accuracy of the classification varies from 70 to 79 per cent depending to the methods of assessment. Causes of errors are numerous: limited spectral information of the optical images, resolution difference between the two data, high density of buildings but most importantly, certain types of building collapses could not be detected by vertically taken images (the case of data in this study). Originality/value – The automatic damage mapping developed in this paper proves to be reliable and could be used in emergency situations. It could also be combined with manual visual interpretation to accelerate the planning of humanitarian rescues and reconstruction.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".