Improved YOLOv5-based multi-crack detection in concrete wall surfaces
Bibliographic record
Abstract
In engineering and infrastructure management, it is essential to ensure the safety and longevity of structures. This can be carried out by properly detecting and monitoring surface cracks. From this lens, this paper presents a novel method for identifying surface cracks using the YOLOv5 (You Only Look Once) deep learning architecture to provide real-time object detection capabilities. First, the YOLOv5 was used with predefined hyperparameters to evaluate its pre-trained knowledge in concrete cracking. Then, the YOLOv5 architecture was fine-tuned to accommodate the specific characteristics of surface cracks within concrete structural components. Finally, the model's backbone was replaced with ResNet-50, and its performance was examined. The experiments in this study involved a diverse dataset of surface crack images and aimed to compare the performance of the three approaches in terms of Precision, Recall, and mean Average Precision metrics. Our findings indicate that YOLOv5-based approaches possess good surface crack identification, with the backbone replacement approach demonstrating the potential for improved adaptability to various structural environments. By combining the capabilities of YOLOv5 and the training strategies, the approach enhances the accuracy and reliability of the surface crack detection systems, resulting in the overall safety and durability of critical infrastructure.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".