viknesh1211/NASA_IMPACT_TCB_OBJ

Transverse Cirrus Bands (TCB) Dataset

Dataset Overview

This dataset contains manually annotated satellite imagery of Transverse Cirrus Bands (TCBs), a type of cloud formation often associated with atmospheric turbulence. The dataset is formatted for object detection tasks using the YOLO and COCO annotation formats, making it suitable for training deep learning models for automated TCB detection.

Data Collection

• Source: NASA-IMPACT Data Share
• Satellite Sensors: VIIRS (Visible Infrared Imaging Radiometer Suite), MODIS (Moderate Resolution Imaging Spectroradiometer)
• Acquisition Method: Downloaded via AWS S3

Annotation Details

• Format: YOLO (.txt) and COCO (.json)
• Bounding Box Labels: Transverse Cirrus Bands (TCB)
• Annotation Tool: MakeSense.ai
• Total Images: X (To be specified)
• Train/Validation/Test Split: 70% / 20% / 10%

File Structure

TCB_Dataset/
│── images/
│   ├── train/
│   ├── val/
│   ├── test/
│── labels/
│   ├── train/
│   ├── val/
│   ├── test/
│── annotations/
│   ├── COCO_format.json
│── README.md

Potential Applications

• Turbulence Detection: Enhancing aviation safety by predicting turbulence-prone regions.
• AI-based Weather Prediction: Training deep learning models for real-time cloud pattern analysis.
• Climate Research: Studying the impact of TCBs on atmospheric dynamics and climate change.
• Satellite-based Hazard Assessment: Detecting and monitoring extreme weather events.

How to Use

1. Clone the repository:

   git clone <repo_link>
   

2. Load images and annotations into your object detection model pipeline.
3. Train models using YOLOv8 or any compatible object detection framework.

Citation

If you use this dataset in your research, please cite:

@article{TCB_Dataset2024,
  title={A Manually Annotated Dataset of Transverse Cirrus Bands for Object Detection in Satellite Imagery},
  author={Your Name},
  year={2024},
  journal={Hugging Face Dataset Repository}
}

License

mit

---

This dataset is open for contributions. Feel free to submit pull requests or raise issues for improvements!