Model Card for rebotnix/rb_productInspection
🏭 Industrial Product Anomaly Detection – Trained by KINEVA, Built by REBOTNIX, Germany Current State: in production and re-training.
This object detection model is specialized in identifying anomalies in manufactured industrial products, such as gears, valves, and machine parts. It has been trained on a curated dataset of common and uncommon industrial defects, the model aids in automated quality control and visual inspection workflows in manufacturing environments.
Developed and maintained by REBOTNIX, Germany, https://rebotnix.com
About KINEVA
KINEVA® is an automated training platform based on the MCP Agent system. It regularly delivers new visual computing models, all developed entirely from scratch. This approach enables the creation of customized models tailored to specific client requirements, which can be retrained and re-released as needed. The platform is particularly suited for applications that demand flexibility, adaptability, and technological precision—such as industrial image processing, smart city analytics, or automated object detection.
KINEVA is continuously evolving to meet the growing demands in the fields of artificial intelligence and machine vision. https://rebotnix.com/en/kineva
✈️ Example Predictions
| Input Image | Detection Result |
|---|---|
 |
 |
 |
 |
| (More example visualizations coming soon) |
Model Details
- Architecture: RF-DETR (custom training head with optimized anchor boxes)
- Task: Object Detection (Anomalies in manufactured products)
- Trained on: REBOTNIX Industrial Defect Dataset (proprietary)
- Format: PyTorch
.pth+ ONNX and trt export available on request - Backbone: EfficientNet B3 (adapted)
- Training Framework: PyTorch + RF-DETR + custom augmentation
Chart
Dataset
The training dataset consists of high-resolution inspection images collected from:
- Partner manufacturing lines
- In-house simulated defect generation
- Custom annotated bounding boxes by REBOTNIX team
The model was trained to be robust across:
- Varying lighting and surface finishes
- Small-scale defects and irregular shapes
- Complex product geometries
Intended Use
| ✅ Intended Use | ❌ Not Intended Use |
|---|---|
| Industrial visual inspection | Consumer photography analysis |
| Quality control automation | Medical imaging |
| Anomaly detection on gears and parts | Non-industrial object detection |
Limitations
- May produce false positives in visually noisy backgrounds
- Performance may degrade on unseen product types not in the training data
- Not trained for thermal or infrared imagery
Usage Example
import supervision as sv
from PIL import Image
from rfdetr import RFDETRBase
model_path= "./rb_productInspection.pth"
CLASS_NAMES = ["anomaly"]
model = RFDETRBase(pretrain_weights=model_path,num_classes=len(CLASS_NAMES))
image_path = "./example_productInspection1.jpg"
image = Image.open(image_path)
detections = model.predict(image, threshold=0.35)
labels = [
f"{CLASS_NAMES[class_id]} {confidence:.2f}"
for class_id, confidence
in zip(detections.class_id, detections.confidence)
]
print(labels)
annotated_image = image.copy()
annotated_image = sv.BoxAnnotator().annotate(annotated_image, detections)
annotated_image = sv.LabelAnnotator().annotate(annotated_image, detections, labels)
annotated_image.save("./output_1.jpg")
Contact
📫 For commercial use or re-training this model support, or dataset access, contact:
REBOTNIX
✉️ Email: [email protected]
🌐 Website: https://rebotnix.com
License
This model is released under CC-BY-NC-SA unless otherwise noted. For commercial licensing, please reach out to the contact email.