A novel method for monitoring material behaviour in urban buildings using aerial multi-modality imaging
Bibliographic record
Abstract
Thermal bridges represent critical weaknesses in building envelope materials, increasing energy consumption by 40% and reducing occupant comfort. This study proposes a novel non-destructive methodology for monitoring material performance in urban buildings using aerial multi-modality imaging. Two models were developed: 1) a customized deep learning (DL) model based on YOLOv9, trained on a multimodal dataset of 5,614 visible, thermal, and LiDAR data, and 2) a physics-based heat loss model for quantifying thermal bridges. The DL model achieved a precision of 0.72, detecting an average of 3.33 thermal bridges per image, with an inference time under 30.3 milliseconds on an NVIDIA A100 GPU, enabling real-time city-scale diagnostics. Post-processing using an ANN refined bounding box predictions and increased precision to 0.763 and reduced localization error by 14.7%. The heat loss model estimated surface losses ranging from 4.02 to 37.85 W/m², with an average of 22.37 W/m². A sensitivity analysis revealed that detection errors caused up to 28.55% relative error in heat loss estimates, with missed detections having the largest impact on performance. Validation on the AGAP (RGB and thermal fused) dataset confirmed generalizability, achieving 80% precision. Comparative evaluation showed YOLOv9-E outperformed other state-of-the-art models such as MaskRCNN, and YOLOv7. Projections suggest that undetected bridges on a typical 100 m² façade may lead to up to 11,409 kWh of excess heating demand over six months. This integrated solution offers a scalable, automated framework for non-destructive testing (NDT), material-level diagnostics, and energy-efficient retrofitting in smart city applications.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 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".