rfdetr-pid-detector

---

RF-DETR P&ID Symbol Detector

A fine-tuned RF-DETR model for detecting Process & Instrumentation Diagram (P&ID) symbols.

The model was developed as part of the paper:

Towards Automated P&ID Digitization: Graph-Based OCR Consolidation and Global Symbol–Tag Association
Accepted at ACM Symposium on Document Engineering (DocEng 2026).

The detector recognizes the graphical symbols appearing in industrial P&IDs and is intended as the first stage of a complete P&ID digitization pipeline.

---

Training

The model was fine-tuned for 10 epochs on a custom P&ID symbol dataset containing 32 symbol classes. During training, both the base model and its Exponential Moving Average (EMA) weights were monitored. The EMA model was selected as the final checkpoint because it consistently achieved higher detection performance.

<p align="center"> <img src="metrics_plot.png" width="100%"> </p>

---

Detected Classes

The model detects the following classes:

---

Installation

pip install rfdetr supervision

For tiled inference:

pip install sahi

---

Load the model

from rfdetr import RFDETRBase

model = RFDETRBase(
    pretrain_weights="checkpoint_best_total.pth"
)

---

Inference (without SAHI)

import cv2
import supervision as sv

image = cv2.imread("image.png")

detections = model.predict(
    image,
    threshold=0.5
)

labels = [
    f"{CLASS_NAMES[c]} {conf:.2f}"
    for c, conf in zip(
        detections.class_id,
        detections.confidence
    )
]

annotated = image.copy()

annotated = sv.BoxAnnotator().annotate(
    annotated,
    detections
)

annotated = sv.LabelAnnotator().annotate(
    annotated,
    detections,
    labels
)

sv.plot_image(annotated)

---

Inference using SAHI

For very large engineering drawings (typically PDF pages rendered at high resolution), tiled inference significantly improves recall.

Recommended parameters:

• Slice size: 1280 × 1280
• Overlap: 20%

from sahi import AutoDetectionModel
from sahi.predict import get_sliced_prediction

detection_model = AutoDetectionModel.from_pretrained(
    model_type="roboflow",
    model=model,
    confidence_threshold=0.5,
    category_mapping=CLASS_NAMES,
    device="cuda",
)

result = get_sliced_prediction(
    image,
    detection_model=detection_model,
    slice_height=1280,
    slice_width=1280,
    overlap_height_ratio=0.2,
    overlap_width_ratio=0.2,
)

The resulting detections are available in

result.object_prediction_list

or can be converted into Supervision detections for visualization.

---

Example

Without SAHI:

Large drawings may miss small symbols.

With SAHI:

Large drawings are processed tile-by-tile, improving the detection of small symbols and densely packed regions.

---

📊 Test Performance

Overall Performance

---

Per-Class Performance

---

🚀 Highlights

• 33 P&ID symbol classes
• mAP@50: 99.96%
• mAP@50:95: 97.89%
• Precision: 99.97%
• Recall: 99.00%
• Control Valve achieved perfect detection performance (100% mAP).
• More than 85% of the classes achieved a mAP@50:95 greater than 98%.
• The EMA model consistently outperformed the base model during training and was selected as the final released checkpoint.
• Optimized for high-resolution P&ID drawings and compatible with SAHI for sliced inference on large engineering diagrams.

---

Citation

If you use this model in your research, please cite:

@inproceedings{XXXX,
  title={Towards Automated P\&ID Digitization: Graph-Based OCR Consolidation and Global Symbol--Tag Association},
  author={...},
  booktitle={Proceedings of the ACM Symposium on Document Engineering (DocEng)},
  year={2026}
}

---

Acknowledgements

This model is built upon the excellent RF-DETR object detector and supports tiled inference through SAHI.

• RF-DETR: https://github.com/roboflow/rf-detr
• SAHI: https://github.com/obss/sahi

---

License

Please refer to the license accompanying this repository.