FIRE DETECTION AND ILLEGAL ACTION SOUNDS DETECTION USING DEEP LEARNING IN DEEP FOREST

This project can be used to detect fire occurance in deep forest through computer vision and detects chainsaw sounds which are illegal felling in the deep forest using deep learning.

Download our Model from : LINK

Download our Demo from : LINK

Setup

  Once you downloaded the model from the gdriver , Place the model folder inside our teamstartup folder.
  Please make sure you have Anaconda3 setup in your computer to have conda environment 

COMMANDS

 1. Open anaconda3 powershell
 2. Navigate to the folder (teamStartup)
 3. conda create --name your_environment_name
 4. conda activate your_environment_name
 5. conda install pip
 6. pip install -r requirements.txt
 7. pip install pipwin
 8. pipwin install pyaudio

Run

  Once you activated and installed all the packages in the environment its time to start.

  RUN : python.exe team_startup.py 

Running the experiment or detection requires that you have Tensorflow, OpenCV and PyAudio installed.

Alert Email ID

  use this Email ID to check the Alert messages from the system.

  Email ID : teamstartuptesting@gmail.com   
  Psw      : ********

PROJECT DEMO

Reference

YOLO(FIRE DETECTION )

    https://pjreddie.com/darknet/yolo/ 

CHAINSAW

   https://www.fastcompany.com/90435386/this-network-of-microphones-listens-for-the-chainsaws-of-illegal-loggers-in-the-rainforest

Chainsaw

Dataset which we used for chainsaw https://research.google.com/audioset/balanced_train/chainsaw.html

you could you use this youtube video to test our chainsaw Model , This video was the one tested in the Demo clip https://www.youtube.com/watch?v=YznLBbgh_F4