Skip to content

An animal classification system developed using transfer learning with the ResNet50 convolutional neural network pre-trained on ImageNet. Designed to distinguish between three classes of animals—cats, dogs, and snakes—the system demonstrates a high accuracy of approximately 98.67% on a balanced dataset comprising 3,000 images.

License

Notifications You must be signed in to change notification settings

AlvaroVasquezAI/Animal_Image_Classification

Repository files navigation

Animal image classification using transfer learning with ResNet50

This project develops an animal image classification system leveraging the ResNet50 model pre-trained on ImageNet. The classifier distinguishes among three animal categories: cats, dogs, and snakes, with a demonstrated high accuracy on a balanced dataset.

Features

  • Utilizes the ResNet50 model pre-trained on ImageNet for high accuracy.
  • Classifies images into three categories: cats, dogs, and snakes.
  • Includes a custom-built GUI using Tkinter for interactive classification.
  • Employs data augmentation techniques to enhance model generalization.

Installation

  1. Clone the repository: git clone https://github.com/AlvaroVasquezAI/Animal_Image_Classification.git

  2. Navigate to the project directory:

  3. Install the required dependencies: pip install -r requirements.txt

Usage

To run the classifier with the GUI, execute (Make sure you are in the correct directory): python main.py

Follow the GUI prompts to select and classify images.

Dataset

The dataset consists of 3,000 images, evenly distributed among the three categories. Images were resized to 256x256 pixels to facilitate processing.

Methodology

The project follows a systematic approach:

  1. Data preprocessing and augmentation to prepare the dataset.
  2. Loading and fine-tuning the ResNet50 model.
  3. Training the model and evaluating its performance.
  4. Developing a Tkinter-based GUI for real-time classification.

Results

The model achieved an accuracy of approximately 98.67% on the test dataset, showcasing the effectiveness of transfer learning in image classification.

Predictions:

Dog

Cat

Snake

Future Work

  • Expand the dataset to include more animal classes.
  • Explore advanced deep learning models to enhance classification accuracy.
  • Develop a web or mobile application to increase accessibility.

Contributors

  • Álvaro García Vásquez

Contact

About

An animal classification system developed using transfer learning with the ResNet50 convolutional neural network pre-trained on ImageNet. Designed to distinguish between three classes of animals—cats, dogs, and snakes—the system demonstrates a high accuracy of approximately 98.67% on a balanced dataset comprising 3,000 images.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published