Skip to content

Machine learning using a fingerprint dataset. The dataset contains real and altered fingerprints. The altered fingerprints are divided into easy, medium and hard altered. Three machine learning models were created with this dataset. A fingerprint identification , a gender classification and a fingerprint differentiation model (real and altered).

Notifications You must be signed in to change notification settings

Naggita-Ethel/Fingerprint-models

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 

Repository files navigation

Recess Project Template

License

Project Logo

Table of Contents

Introduction

These models were developed using Convolutional Neural Networks (CNN) and ResNet Neural Networks. Various strategies including regularization, data augmentation, early stopping, dropout and random oversampling, were employed to address challenges posed by class imbalance and overfitting. The dataset used was the sokoto conventry fingerprint dataset obtained from kaggle. This contains 6000 real images, 17931 easy altered images, 17067 medium altered images and 14272 hard altered images. Evaluation techniques like the AUC-ROC curve and classification reports were used.

Link to the dataset https://www.kaggle.com/datasets/ruizgara/socofing

Collaboration Process

flowchart LR

subgraph "Repository"
  Repo((Repository))
end

subgraph "User A"
  fork[Fork]
  branchA[Branch A]
end

subgraph "User B"
  branchB[Branch B]
end

subgraph "User C"
  branchC[Branch C]
end

subgraph "Pull Request"
  pullRequest[Pull Request]
end

Repo --> fork
fork --> branchA
fork --> branchB
fork --> branchC

branchA --> pullRequest
branchB --> pullRequest
branchC --> pullRequest

pullRequest --> Repo
Loading

Note: Each Group member has to Own a Branch named as his/her name

Documentation

Find detailed project documentation in the docs folder. It includes:

  • Project Overview: Get an overview of the project's goals, objectives, and how to use this template effectively.

  • Project Guidelines: Learn about the best practices for group collaboration, code organization, and documentation.

  • Reports and Presentations

Submit your final project reports and group presentation in the reports folder:

Resources

The resources folder contains additional resources for the project, including:

  • Images: Store images related to your project (e.g., diagrams, graphics).
  • Datasets: If applicable, keep datasets used in your project for reproducibility.

Getting Started

To begin your recess project, use the "Use this template" button to create your individual project repository based on this template. Follow these steps to get started:

  1. Clone your individual project repository to your local machine.
  2. Collaborate with your group members on the group project within the src/group_work folder.
  3. Work on your individual contributions within your respective folders in src/individual_work.
  4. Regularly commit and push changes to your repository.
  5. Create pull requests for group project changes and get feedback from your team.

❗️ Note:

Contributing

If you want to contribute to this project template or suggest improvements, please follow the guidelines outlined in CONTRIBUTING.md. We welcome your contributions and value your feedback!

License

This project is licensed under the MIT License. Feel free to use and modify this template for your own recess projects.


About

Machine learning using a fingerprint dataset. The dataset contains real and altered fingerprints. The altered fingerprints are divided into easy, medium and hard altered. Three machine learning models were created with this dataset. A fingerprint identification , a gender classification and a fingerprint differentiation model (real and altered).

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •