Colony Counter

Colony Counter icon

Model Description

Colony Counter is an application within the CarbConnect platform. It automatically detects and counts bacterial colonies from agar plate images.

The application returns the total colony count and bounding boxes for detected colonies, with optional annotated output for visualization.

This repository contains the model card for the private Hugging Face model repository associated with the application. A companion Space for this app already exists under the CarbGeM organization.

Intended Use and Limitations

Research Use Only (RUO).

This tool is strictly intended for research use only and is not for use in clinical diagnostics or medical procedures. For further details, please refer to the Colony Counter Learn More page on CarbConnect.

Intended users

  • Microbiologists
  • Researchers
  • Laboratories performing bacterial colony analysis

Out-of-scope use

  • Clinical diagnostics
  • Medical procedures
  • Use as a standalone diagnostic device

Practical limitations

  • Output quality depends on the quality, angle, lighting, and resolution of the agar plate image.
  • Colony overlap, artifacts, reflections, and dense growth patterns may affect detection accuracy.
  • Results are intended to support research workflows and should not replace expert review.

How to Use

Typical workflow:

  1. Capture or upload an image of an agar plate.
  2. Run inference through the Colony Counter application.
  3. Review the total colony count and the detected bounding boxes.
  4. Optionally inspect the annotated output image for visual verification.
  5. Use the result for research support only, not for clinical decision-making.

Training Data

The model was trained using the following dataset:

  • Dataset name: Annotated dataset for deep-learning-based bacterial colony detection
  • Description: A dataset containing 369 digital images of 24 bacteria species cultures of veterinary importance, with a total of 56,865 manually annotated bounding boxes for bacterial colonies
  • Authors: Norbert Solymosi, Sára Nagy
  • Source / DOI: 10.6084/m9.figshare.22022540
  • License: CC BY 4.0
  • Modifications: The dataset was used in its original form without any modifications

Outputs

For each processed image, the application can provide:

  • Total detected colony count
  • Bounding boxes for detected colonies
  • Optional annotated output image

Citation

If you use or reference the training dataset, please cite:

Solymosi, Norbert and Nagy, Sára. Annotated dataset for deep-learning-based bacterial colony detection. figshare. Dataset. 10.6084/m9.figshare.22022540

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support