A multiclassifier convolutional neural network to identify defect type and severity in roofing elements
Bibliographic record
Abstract
Purpose Roofing is highly susceptible to environmental damage from elements like wind, snow and rain. Regular inspection and maintenance are essential to extend a roof’s lifespan. This study aims to develop an automated system that detects and classifies roofing damage types and their severity using image-based analysis, helping asset managers prioritize repairs and allocate maintenance resources more effectively. Design/methodology/approach This study uses Convolutional Neural Networks (CNNs) for image-based damage detection and classification. Over 3,000 images of roofing segments (1.5 × 1.12 m) from institutional buildings were used for training and testing. The model first identifies damage type – no damage, vegetation or ponding – then classifies vegetation damage severity into low, moderate or severe. Findings The developed CNN model achieved over 94% accuracy in both damage type and severity classification. The results demonstrate the model’s effectiveness in analyzing roofing defects. Research limitations/implications Future enhancements include expanding the system to detect additional defect types like cracks and flashing defects, offering a scalable solution for systematic roof condition assessment and maintenance planning. Originality/value Unlike traditional manual inspections, this approach uses computer vision techniques to offer a scalable, data-driven framework that identifies damage types and quantifies severity levels. This makes roofing inspections more efficient, consistent and safer.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".