Scribd
Upload
6 views5 pages

Rescue Sound Gas Documentation

The document presents a rescue signal system designed for detecting human presence and gas leaks in collapsed structures using sound classification and gas detection technologies. It details the system's architecture, which integrates a PC for audio processing with an Arduino-based embedded system for real-time feedback, achieving a classification accuracy of 90%. Future enhancements include transitioning to a Raspberry Pi for processing, adding GPS for location tracking, and expanding sensor capabilities.

Uploaded by

safa.salih
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
6 views5 pages

Rescue Sound Gas Documentation

The document presents a rescue signal system designed for detecting human presence and gas leaks in collapsed structures using sound classification and gas detection technologies. It details the system's architecture, which integrates a PC for audio processing with an Arduino-based embedded system for real-time feedback, achieving a classification accuracy of 90%. Future enhancements include transitioning to a Raspberry Pi for processing, adding GPS for location tracking, and expanding sensor capabilities.

Uploaded by

safa.salih
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Rescue Signal System: Intelligent

Detection of Human Presence and Gas


Leaks in Collapsed Structures
Submitted by: Your Name Here

Date: April 2025


Table of Contents
1. 1. Introduction
2. 2. Background and Literature Review
3. 3. Solution Design
4. 4. Data Analysis and Results
5. 5. Conclusion and Future Work
6. 6. References
7. 7. Appendix
1. Introduction
Natural disasters such as earthquakes often result in victims trapped beneath collapsed
buildings. Rapid identification of these individuals is crucial for increasing survival rates.
This project introduces an innovative system that leverages sound classification and gas
detection to support search-and-rescue missions. The solution combines machine learning
on a PC with real-time feedback on an Arduino-based embedded system.

2. Background and Literature Review


During initial research, I sought a project centered around identifying human sounds under
rubble. While several studies exist on drone and thermal imaging in rescue, very few
targeted real-time audio-based detection integrated with gas sensing.

Early trials involved using Arduino-compatible microphones. However, these sensors


lacked the sensitivity needed to distinguish complex human sounds. Thus, the approach
evolved: using a laptop to record and classify audio using Python, then sending results to
the Arduino via serial communication.

To enhance the system, a gas sensor (MQ-2) was added, providing alerts when dangerous
gas levels were present. This addition is vital for rescuers' safety and can inform decisions
like wearing masks or relocating victims.

Scientific References:
1. D. L. Donoho, "De-noising by soft-thresholding", IEEE Transactions on Information
Theory, 1995.
2. L. R. Rabiner, "A tutorial on hidden Markov models and selected applications in speech
recognition", Proc. IEEE, 1989.
3. MQ-2 Datasheet – Hanwei Electronics.
4. R. A. Jarvis and A. A. Byrne, "Search and rescue robotics – A review", Elsevier Robotics and
Autonomous Systems, 2003.
5. T. Virtanen et al., "Computational Analysis of Sound Scenes and Events", Springer, 2018.

3. Solution Design
System Architecture

- PC Side:
- Records audio using `sounddevice`.
- Extracts MFCC features using `librosa`.
- Classifies using a pre-trained SVM model.
- Sends result ('G' for people, 'R' for noise/animal) to Arduino.
- Arduino Side:
- Displays gas value and sound classification on a 20x4 LCD.
- Lights up green (human detected), red (noise), or yellow (dangerous gas).
- Button used to reset LEDs.

Components Used:

| Component | Function |
|--------------------------|--------------------------------------------------------|
| Arduino Uno | Core microcontroller |
| MQ-2 Gas Sensor | Detects smoke, propane, methane, alcohol, LPG |
| LCD 20x4 I2C | Displays gas and sound info |
| LEDs (Green/Red/Yellow) | Visual indicators (human, noise, gas alert) |
| 220Ω Resistors | Current limiting for LEDs |
| Push Button | Resets alert state |
| Breadboard & Jumper Wires| Circuit prototyping |

Gas Levels Based on MQ-2:

| Sensor Value (Analog) | Gas Level Description | Action |


|------------------------|-----------------------------|----------------------|
| 0–200 | Clean Air | No alert |
| 200–400 | Light Gas Presence | Monitor only |
| 400–600 | Medium Gas Level | Caution |
| >600 | Dangerous Gas Level | Yellow LED ON |

4. Data Analysis and Results


Audio Dataset

- Total Samples: 268 WAV files


- Classes: "People", "Animal and noise"
- Features: 20 MFCC coefficients averaged per sample
- Model: SVM with linear kernel
- Accuracy: 90.0% on 80/20 train-test split

Example Output

The system is able to:


- Detect human presence and trigger a green LED
- Detect animal/background noise and trigger a red LED
- Detect hazardous gas concentration and trigger a yellow LED
- Show real-time classification and gas values on LCD
5. Conclusion and Future Work
This project presents a robust, low-cost tool for disaster response. It successfully combines
sound classification with environmental sensing, using affordable components.

Future Enhancements

- Transition from Arduino to Raspberry Pi for onboard audio processing


- Add GPS module to geolocate detection sites
- Integrate GSM for SMS alerts to rescue teams
- Expand sensors (CO, CO₂, temperature)
- Enhance audio model to detect whispers and distress tones

6. References
See section 2: Background and Literature Review

7. Appendix
A. Arduino Code
[ARDUINO CODE PLACEHOLDER]

B. Python: Model Training Code


[TRAIN_MODEL.PY CODE PLACEHOLDER]

C. Python: Live Detection Code


[LIVE_DETECTION.PY CODE PLACEHOLDER]

D. Project Title Suggestions:


1. VocalGas Rescue
2. EchoSafe
3. SAR-Sense
4. Human Echo Finder
5. SafeSignal
6. ResQ Intel
7. SoundGas Guardian

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy