|
|
--- |
|
|
license: mit |
|
|
task_categories: |
|
|
- object-detection |
|
|
- image-segmentation |
|
|
tags: |
|
|
- yolo |
|
|
- railway |
|
|
- security |
|
|
- surveillance |
|
|
- yolov8 |
|
|
pretty_name: RailRakshak Security Dataset |
|
|
size_categories: |
|
|
- 1K<n<10K |
|
|
--- |
|
|
|
|
|
# 🚄 RailRakshak: Railway Threat Detection Dataset |
|
|
|
|
|
 |
|
|
 |
|
|
|
|
|
## 📖 Dataset Description |
|
|
|
|
|
|
|
|
### 🎯 Supported Tasks |
|
|
- **Track Segmentation:** Pixel-wise masks of the railway track (Safe Zone). |
|
|
- **Threat Detection:** Bounding boxes for obstacles on the track. |
|
|
|
|
|
## 📂 Dataset Structure |
|
|
Due to the large number of small files, the dataset is provided as a **single ZIP archive** (`dataset.zip`) to ensure fast downloads and bypass Git LFS rate limits. |
|
|
|
|
|
**⚠️ You must unzip the file after downloading.** |
|
|
|
|
|
### Directory Layout (After Unzipping) |
|
|
```bash |
|
|
dataset/ |
|
|
├── railway_dataset/ |
|
|
│ ├──Rail-DB/ |
|
|
├── railway_seg |
|
|
│ ├── train/ |
|
|
│ └── raildb_raw/ |
|
|
├── samples/ |
|
|
│ ├── test.mp4 |
|
|
│ └── Test2.mp4 |
|
|
└── output/ # EMPTY DIRECTORY |
|
|
|
|
|
``` |
|
|
|
|
|
## 🚀 How to Use |
|
|
|
|
|
### Option 1: Direct Download |
|
|
|
|
|
1. Go to the **"Files and versions"** tab. |
|
|
2. Download `dataset.zip`. |
|
|
3. Unzip it locally: |
|
|
```bash |
|
|
unzip dataset.zip |
|
|
|
|
|
``` |
|
|
|
|
|
### Option 2: Python (Hugging Face Hub) |
|
|
|
|
|
You can download and unzip programmatically using Python: |
|
|
|
|
|
```python |
|
|
from huggingface_hub import hf_hub_download |
|
|
import zipfile |
|
|
|
|
|
# Download |
|
|
path = hf_hub_download( |
|
|
repo_id="YOUR_USERNAME/RailRakshak-Track-Detection-Data", |
|
|
filename="dataset.zip", |
|
|
repo_type="dataset" |
|
|
) |
|
|
|
|
|
# Unzip |
|
|
with zipfile.ZipFile(path, 'r') as zip_ref: |
|
|
zip_ref.extractall("./railrakshak_data") |
|
|
|
|
|
print("✅ Dataset downloaded and extracted!") |
|
|
|
|
|
``` |
|
|
|
|
|
### Option 3: Training with YOLOv8 |
|
|
|
|
|
Once unzipped, you can train immediately: |
|
|
|
|
|
```bash |
|
|
yolo task=segment mode=train model=yolov8n-seg.pt data=./railrakshak_data/data.yaml epochs=100 imgsz=640 |
|
|
|
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
## ✍️ Citation |
|
|
|
|
|
If you use this dataset, please cite the **RailRakshak Hackathon Project**. |
|
|
|
|
|
*Created by [RAT]* |
|
|
|
|
|
--- |
|
|
|
