Link to this sectionSignature Detection Dataset#

The Ultralytics Signature Detection Dataset is an object detection dataset of 178 document images annotated with a single signature class, pre-split into 143 training and 35 validation images. The dataset downloads automatically (11.3 MB) the first time you train, making it a compact starting point for computer vision applications such as document verification, fraud detection, and digital document processing.

Link to this sectionDataset Structure#

The dataset contains 178 images of various document types with handwritten signatures, split into two subsets:

Every image carries bounding-box annotations for one class, signature, and the configuration defines no separate test split.

Explore Signature on Ultralytics Platform to browse the images with their annotation overlays, view the class distribution and bounding-box heatmaps in the Charts tab, and clone it to train your own model in the cloud.

Link to this sectionApplications#

A model trained on this dataset can identify and track signatures in scanned documents and video, supporting:

• Document Verification: Automating signature checks in legal and financial documents
• Fraud Detection: Identifying potentially forged or unauthorized signatures
• Digital Document Processing: Streamlining workflows in administrative and legal sectors
• Banking and Finance: Enhancing security in check processing and loan document verification
• Archival Research: Supporting historical document analysis and cataloging
• Education and Research: Studying signature characteristics across document types in computer vision courses

Link to this sectionDataset YAML#

The signature.yaml file defines the dataset configuration — the dataset paths, class names, and other metadata. It is maintained in the Ultralytics repository at https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/datasets/signature.yaml.

# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license

# Signature dataset by Ultralytics
# Documentation: https://docs.ultralytics.com/datasets/detect/signature
# Example usage: yolo train data=signature.yaml
# parent
# ├── ultralytics
# └── datasets
#     └── signature ← downloads here (11.3 MB)

# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: signature # dataset root dir
train: images/train # train images (relative to 'path') 143 images
val: images/val # val images (relative to 'path') 35 images

# Classes
names:
  0: signature

# Download script/URL (optional)
download: https://github.com/ultralytics/assets/releases/download/v0.0.0/signature.zip

Link to this sectionUsage#

To train a YOLO26n model on the Signature Detection Dataset for 100 epochs with an image size of 640, use the provided code samples. For a comprehensive list of available parameters, refer to the model's Training page.

from ultralytics import YOLO

# Load a model
model = YOLO("yolo26n.pt")  # load a pretrained model (recommended for training)

# Train the model
results = model.train(data="signature.yaml", epochs=100, imgsz=640)

Once trained, you can run inference on documents or video with the fine-tuned model. The example below runs prediction on a sample video with a confidence threshold of 0.75:

from ultralytics import YOLO

# Load a model
model = YOLO("path/to/best.pt")  # load a signature-detection fine-tuned model

# Inference using the model
results = model.predict("https://ultralytics.com/assets/signature-s.mp4", conf=0.75)

Link to this sectionSample Images and Annotations#

The dataset covers a variety of document formats, helping trained models generalize across contracts, forms, and letters. Below is a training batch from the dataset:

• Mosaiced Image: Here, we present a training batch consisting of mosaiced dataset images. Mosaicing, a training technique, combines multiple images into one, enriching batch diversity. This method helps enhance the model's ability to generalize across different signature sizes, aspect ratios, and contexts.

Link to this sectionCitations and Acknowledgments#

The dataset has been made available under the AGPL-3.0 License.

If you use the Signature Detection Dataset in your research or development work, please cite it appropriately:

@dataset{Ultralytics_Signature_Detection_Dataset_2024,
    author = {Ultralytics},
    title = {Signature Detection Dataset},
    year = {2024},
    publisher = {Ultralytics},
    url = {https://docs.ultralytics.com/datasets/detect/signature/}
}

Link to this sectionFAQ#

Link to this sectionWhat is the Signature Detection Dataset used for?#

The Signature Detection Dataset is a collection of 178 annotated document images for training models to detect handwritten signatures. It supports document verification, fraud detection, and archival research, and is a practical base for building smart document analysis systems with machine learning.

Link to this sectionHow do I download the Signature Detection Dataset?#

The dataset downloads automatically (11.3 MB) from Ultralytics GitHub assets the first time you train with data="signature.yaml" — no manual download is required. To explore other datasets, browse the detection datasets overview.

Link to this sectionHow many images and classes are in the Signature Detection Dataset?#

The Signature Detection Dataset contains 143 training and 35 validation images — 178 in total — each annotated with a single class, signature. There is no separate test split. See the Dataset Structure section and the signature.yaml configuration for details.

Link to this sectionHow do I train a YOLO26n model on the Signature Detection Dataset?#

You can train a YOLO26n model for 100 epochs with an image size of 640 using Python or the CLI:

from ultralytics import YOLO

# Load a pretrained model
model = YOLO("yolo26n.pt")

# Train the model
results = model.train(data="signature.yaml", epochs=100, imgsz=640)

For more details, refer to the Training page and model training tips.

Link to this sectionHow can I run inference with a model trained on the Signature Detection Dataset?#

Load your fine-tuned weights and run prediction:

from ultralytics import YOLO

# Load the fine-tuned model
model = YOLO("path/to/best.pt")

# Perform inference
results = model.predict("https://ultralytics.com/assets/signature-s.mp4", conf=0.75)

Link to this sectionCan I use the Signature Detection Dataset in commercial projects?#

The dataset is released under the AGPL-3.0 License, which permits commercial use provided derivative works — including software offered over a network — are made available under the same license. For licensing options that remove the open-source requirements, see Ultralytics Licensing.