🌫️ SmokeSimulation

Virtual Environment for Synthetic Smoke Cloud Generation

MATLAB source code for synthetic smoke plume generation and wildfire smoke detection research.

---

🔥 Overview

SmokeSimulation provides a MATLAB implementation for generating synthetic smoke clouds over real background videos.

The code was developed for research on virtual smoke plume generation and its use in improving wildfire smoke detection systems. Synthetic smoke allows researchers to enrich video datasets with controlled smoke examples, helping evaluate and train computer vision and computational intelligence methods when real wildfire footage is limited.

This repository accompanies the IEEE VECIMS 2011 paper “Virtual environment for synthetic smoke clouds generation” and is also related to the IEEE TSMC:S 2013 work “Wildfire smoke detection using computational intelligence techniques enhanced with synthetic smoke plume generation.”

---

✨ Key Features

• 🌫️ Synthetic smoke plume generation on real video backgrounds
• 🎥 Video-based simulation workflow using base video sequences
• 🧪 Useful for dataset augmentation in wildfire smoke detection research
• 🧠 Designed to support computational-intelligence smoke detection pipelines
• 🛠️ Lightweight MATLAB implementation
• 📚 Research code associated with peer-reviewed IEEE publications

---

🧭 Simulation Pipeline

Conceptually, the workflow is:

Base video
   │
   ▼
Background frame extraction
   │
   ▼
Synthetic smoke cloud modeling
   │
   ▼
Smoke plume rendering over video frames
   │
   ▼
Synthetic smoke video output
   │
   ▼
Evaluation / training of wildfire smoke detection methods

---

📁 Repository Structure

SmokeSimulation/
│
├── launch_smokeSimulation_base.m   # Main MATLAB script
├── util/                           # Utility functions
├── vids_base/                      # Base videos on which smoke is simulated
├── README.md                       # Project documentation
└── LICENSE                         # GPL-3.0 license

---

🚀 Getting Started

1. Clone the repository

git clone https://github.com/AngeloUNIMI/SmokeSimulation.git
cd SmokeSimulation

2. Add base videos

Place the background videos to be augmented inside:

./vids_base/

These videos are used as the base scenes on which synthetic smoke clouds are rendered.

3. Run the simulation

Open MATLAB, move to the repository folder, and run:

launch_smokeSimulation_base

---

🧪 Typical Use Cases

Use case	Description
Dataset augmentation	Generate synthetic smoke samples from clean or low-smoke videos
Smoke detector testing	Evaluate smoke-detection algorithms under controlled plume conditions
Wildfire monitoring research	Simulate visual smoke events for early detection studies
Reproducible experiments	Recreate the synthetic smoke generation workflow described in the papers

---

📊 Inputs and Outputs

Input

• Base video sequences stored in ./vids_base/
• MATLAB parameters defined in the main script and utility functions

Output

• Video frames or sequences with synthetic smoke clouds rendered over the original background
• Data suitable for smoke detection experiments and visual inspection

The exact output location and naming may depend on the parameters set inside launch_smokeSimulation_base.m.

---

📚 Papers

This code is associated with the following publications:

VECIMS 2011

A. Genovese, R. Donida Labati, V. Piuri, and F. Scotti
Virtual environment for synthetic smoke clouds generation
IEEE International Conference on Virtual Environments, Human-Computer Interfaces and Measurement Systems, Ottawa, Canada, September 2011, pp. 1–6.
DOI: 10.1109/VECIMS.2011.6053841

IEEE TSMC:S 2013

R. Donida Labati, A. Genovese, V. Piuri, and F. Scotti
Wildfire smoke detection using computational intelligence techniques enhanced with synthetic smoke plume generation
IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 43, no. 4, July 2013, pp. 1003–1012.
DOI: 10.1109/TSMCA.2012.2224335

---

🔗 Links

Resource	Link
Project page	https://homes.di.unimi.it/genovese/wild/wildfire.htm
VECIMS 2011 paper	https://ieeexplore.ieee.org/document/6053841
IEEE TSMC:S 2013 paper	https://ieeexplore.ieee.org/document/6425498
Related detection code	https://github.com/AngeloUNIMI/WildfireSmokeDetection

---

🧩 Related Work

The code implements or adapts concepts from smoke-cloud simulation methods described in:

V. N. Vasyukov and A. N. Podovinnikov
Simulating 2D images of smoke clouds for the purpose of fire detection algorithms adjustment
2008 Third International Forum on Strategic Technologies, Novosibirsk-Tomsk, 2008, pp. 369–370.
DOI: 10.1109/IFOST.2008.4602977

---

🏛 Authors

Angelo Genovese, Ruggero Donida Labati, Vincenzo Piuri, and Fabio Scotti

Department of Computer Science, Università degli Studi di Milano, Italy

---

📖 Citation

If you use this code, please cite:

@INPROCEEDINGS{6053841,
  author    = {A. {Genovese} and R. {Donida Labati} and V. {Piuri} and F. {Scotti}},
  booktitle = {2011 IEEE International Conference on Virtual Environments, Human-Computer Interfaces and Measurement Systems},
  title     = {Virtual environment for synthetic smoke clouds generation},
  year      = {2011},
  pages     = {1-6},
  doi       = {10.1109/VECIMS.2011.6053841},
  ISSN      = {1944-9429},
  month     = {Sep.}
}

@Article{tsmca12,
  author  = {R. {Donida Labati} and A. Genovese and V. Piuri and F. Scotti},
  title   = {Wildfire smoke detection using computational intelligence techniques enhanced with synthetic smoke plume generation},
  journal = {IEEE Transactions on Systems, Man and Cybernetics: Systems},
  volume  = {43},
  number  = {4},
  pages   = {1003-1012},
  month   = {July},
  year    = {2013},
  note    = {2168-2216}
}

---

📄 License

This project is released under the GNU General Public License v3.0.

See the LICENSE file for details.