https://github.com/protyayofficial/convsfnet

Enhanced disaster image classification using ConvNeXt with Squeeze-and-Excitation (SE) and Feature Pyramid Network (FPN). Built on the MEDIC dataset, this project aims to improve classification accuracy and address overfitting issues.
https://github.com/protyayofficial/convsfnet

computer-vision deep-learning disaster-identification

Last synced: about 1 year ago
JSON representation

Enhanced disaster image classification using ConvNeXt with Squeeze-and-Excitation (SE) and Feature Pyramid Network (FPN). Built on the MEDIC dataset, this project aims to improve classification accuracy and address overfitting issues.

• Host: GitHub
• URL: https://github.com/protyayofficial/convsfnet
• Owner: protyayofficial
• License: mit
• Created: 2024-07-08T11:16:05.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-07-25T07:40:11.000Z (almost 2 years ago)
• Last Synced: 2025-06-10T22:51:25.426Z (about 1 year ago)
• Topics: computer-vision, deep-learning, disaster-identification
• Language: Python
• Homepage:
• Size: 153 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ConvSFNet: Enhanced Disaster Image Classification

This project enhances disaster image classification by building upon the [Medic repository](https://github.com/firojalam/medic). We introduce a novel architecture combining ConvNeXt with Squeeze-and-Excitation (SE) and Feature Pyramid Network (FPN) to improve classification accuracy and address overfitting issues.

## Directory Structure 

![directory_structure](directory_structure.png)

## Features

- **Enhanced Model Architecture**: Integration of ConvNeXt with Squeeze-and-Excitation (SE) and Feature Pyramid Network (FPN).

- **Improved Preprocessing**: Advanced preprocessing techniques to enhance model performance.

- **Comprehensive Evaluation**: Detailed evaluation metrics and results for various models.

## Download the Dataset

To download the dataset: https://crisisnlp.qcri.org/data/medic/MEDIC.tar.gz

More details about the dataset: https://crisisnlp.qcri.org/medic/

Kindly give proper citation to the original authors

## Acknowledgments

We would like to thank the authors of the Medic repository for providing a solid foundation for our work. Their initial framework was essential in developing our enhanced model.

## License

This project is licensed under the MIT License.