A Flutter application that uses machine learning to classify weather conditions from images. This project was developed for the IndabaX South Sudan 2025 AI Hackathon.

https://drive.google.com/file/d/1HTNr5HVNoRj2IgWpDywXbUcbkAR5ft7r/view?usp=sharing

• Weather Classification: Identify weather conditions (Cloudy, Rain, Shine, Sunrise) from images
• Camera Integration: Take photos directly within the app
• Gallery Selection: Choose existing images from your device
• Real-time Predictions: Get instant classification results
• Confidence Scores: View confidence levels for each weather category
• Modern UI: Beautiful, intuitive interface with smooth animations

• Flutter: Cross-platform UI framework
• TensorFlow Lite: On-device machine learning
• Camera API: Device camera integration
• Image Processing: Image manipulation and preprocessing

The app uses a TensorFlow Lite model trained to classify weather conditions into four categories:

• Cloudy
• Rain
• Shine
• Sunrise

• Flutter SDK (3.8.0 or higher)
• Android Studio / Xcode
• A physical device or emulator

1. Clone the repository:

   git clone https://github.com/yourusername/weather_classifier.git

2. Navigate to the project directory:

   cd weather_classifier

3. Install dependencies:

   flutter pub get

4. Run the app:

   flutter run

flutter build apk

The APK file will be located at build/app/outputs/flutter-apk/app-release.apk

flutter build ios

Then use Xcode to archive and distribute the app.

• lib/: Contains all Dart code
  • main.dart: Entry point of the application
  • splash_screen.dart: Initial loading screen
  • home_page.dart: Main screen with camera and gallery options
  • camera_screen.dart: Camera interface
  • model_service.dart: ML model integration
  • results_page.dart: Display classification results
• assets/: Contains images and model files
  • models/: Contains the TensorFlow Lite model
  • weather_icon.png: App icon and images

This project was developed as part of the IndabaX South Sudan 2025 AI Hackathon, which focuses on applying artificial intelligence solutions to local challenges. The Weather Classifier app demonstrates how machine learning can be used for environmental monitoring and weather prediction in regions where traditional meteorological infrastructure may be limited.

Access Denied Team

• Chol Daniel
• Kenisa Majhok
• Nhial Majok
• Ajang Chol

African Leadership University

This project is licensed under the MIT License - see the LICENSE file for details.

• TensorFlow team for the machine learning framework
• Flutter team for the UI framework
• IndabaX South Sudan for organizing the hackathon