| --- |
| language: |
| - en |
| pipeline_tag: object-detection |
| library_name: ultralytics |
| tags: |
| - industrial-vision |
| - factoryvision-ai |
| - ppe-detection |
| - education |
| - yolo11 |
| --- |
| <div align="center"> |
| <img src="hero.svg" alt="FactoryVision AI PPE detection: YOLO11n object detection for safety equipment" width="100%"> |
| <p><strong>Educational prototype</strong> | Object detection for factory and construction imagery | Human review required</p> |
| </div> |
|
|
| <h2>Model overview</h2> |
| <table> |
| <thead><tr><th>Model</th><th>Task</th><th>Framework</th><th>Weights</th></tr></thead> |
| <tbody><tr><td>FactoryVision AI PPE YOLO11n</td><td>Object detection</td><td>Ultralytics YOLO11n</td><td><code>model.pt</code></td></tr></tbody> |
| </table> |
|
|
| <h2>Held-out test results</h2> |
| <table> |
| <thead><tr><th>Metric</th><th>Reported result</th></tr></thead> |
| <tbody> |
| <tr><td>Precision (bounding boxes)</td><td><strong>0.8400</strong></td></tr> |
| <tr><td>Recall (bounding boxes)</td><td><strong>0.6398</strong></td></tr> |
| <tr><td>mAP@50</td><td><strong>0.7053</strong></td></tr> |
| <tr><td>mAP@50:95</td><td><strong>0.3927</strong></td></tr> |
| </tbody> |
| </table> |
| <p>These values are reported in <code>ppe_test_metrics.json</code>, not independently reproduced. The reported test split has 82 images; a class-specific breakdown is not included in this package. Do not confuse the separate pretrained YOLO COCO person baseline with this PPE result.</p> |
| |
| <h2>Purpose and safety</h2> |
| <p>Educational object detection for people and PPE-related objects in factory and construction images. <strong>This is not a certified industrial safety system.</strong> Detected PPE status is only a suggested alert requiring human review; it cannot establish a worker's actual PPE compliance or substitute for a trained safety inspector. Missed detections mean unsafe situations may go unflagged.</p> |
| <blockquote><strong>Safety limitation:</strong> Object detections alone do not perform the project's person-to-helmet association. The original application implements an additional alert logic layer; consult its source and use human review.</blockquote> |
|
|
| <h2>Model details</h2> |
| <ul> |
| <li>Fine-tuned Ultralytics YOLO11n; original starting checkpoint <code>yolo11n.pt</code></li> |
| <li>Packaged weights: <code>model.pt</code> (archive path <code>checkpoints/ppe_yolo.pt</code>)</li> |
| <li>10 classes: see <code>labels.json</code></li> |
| <li>Training configuration: 30 epochs, 640px image size, batch size 16; the training script uses a held-out PPE test split</li> |
| </ul> |
| |
| <h2>Training data</h2> |
| <p>Construction Site Safety Image Dataset, attributed to Snehil Sanyal / Roboflow Universe and referenced by the source project: <a href="https://www.kaggle.com/datasets/snehilsanyal/construction-site-safety-image-dataset-roboflow">Kaggle dataset</a>. The source project lists CC BY 4.0 for this dataset; verify provenance and attribution requirements before public release. Dataset images are not included in this model package.</p> |
| |
| <h2>Example inference</h2> |
| <pre><code>from ultralytics import YOLO |
| model = YOLO('model.pt') |
| results = model.predict('your_image.jpg', conf=0.25) |
| for result in results: |
| print(result.boxes)</code></pre> |
| |
| <h2>Licensing and redistribution</h2> |
| <ul> |
| <li>Ultralytics YOLO11 and its fine-tuned checkpoints have AGPL-3.0 obligations by default, or require a suitable enterprise licence for uses incompatible with AGPL-3.0. See <a href="https://www.ultralytics.com/license">Ultralytics licensing</a>.</li> |
| <li>The source repository's MIT licence does <strong>not</strong> automatically license this trained checkpoint, pretrained weights, or source datasets.</li> |
| <li>The original developer has reportedly authorised publication; the publisher must still confirm the specific publication terms and applicable third-party rights.</li> |
| <li><strong>Do not assign an MIT/Apache model licence or switch to public visibility until licensing has been checked and recorded.</strong></li> |
| </ul> |
|
|
| <h2>Known limitations</h2> |
| <p>Domain shift, lighting, small objects, occlusion, crowded scenes, cameras and inference thresholds may reduce performance. False negatives are especially concerning for PPE alerting.</p> |
|
|
| <hr> |
| <h2>Credits and attribution</h2> |
| <table> |
| <thead><tr><th>Role</th><th>Contributor</th></tr></thead> |
| <tbody> |
| <tr><td>Original Developer & Model Author</td><td>Theivanth Rajamohan</td></tr> |
| <tr><td>Technical Reviewer, Code Contributor & Publication Support</td><td><a href="https://huggingface.co/Swapinvidya">Swapin Vidya</a></td></tr> |
| <tr><td>Project Contributor</td><td><a href="https://huggingface.co/rajahshan">Rajah Shanmugasundaram</a></td></tr> |
| <tr><td>Publishing Organisation</td><td>TechSphere Consultancy Services Private Limited</td></tr> |
| </tbody> |
| </table> |
| <p><strong>Project provenance:</strong> FactoryVision AI was originally developed by Theivanth Rajamohan. <a href="https://huggingface.co/Swapinvidya">Swapin Vidya</a> contributed technical review, detection-metrics improvements, testing, model publication preparation and documentation. The models are being prepared for publication under TechSphere Consultancy Services Private Limited's Hugging Face namespace, with the original developer's permission.</p> |
| <ul> |
| <li><a href="https://huggingface.co/TechsphereAI">TechSphereAI on Hugging Face</a></li> |
| <li><a href="https://techsphereconsultancy.com/">Official website</a></li> |
| <li><a href="https://github.com/Theivanth/Factory-Vision-AI">Original GitHub project</a></li> |
| <li><a href="https://github.com/Theivanth/Factory-Vision-AI/pull/1">Swapin Vidya's merged contribution</a></li> |
| </ul> |
| |
