Chainsaw sound visualization
Rain sound visualization
Work-Flow
Results (probability)

THE CHALLENGE I HAVE TAKEN

Stop Illegal logging
==> What is Illegal logging ?
==> Illegal logging (major contribution to deforestation) is a serious problem in many places, particularly Indonesia, Cambodia and some parts of Africa. The loggers, often employed by rogue companies, cut down trees in violation of national conservation laws and outside the control of governments.
==> How big a problem is illegal logging ?
==> illegal logging accounts for between 15% and 30% of global timber trade, and rises to 50% to 90% of the trade from tropical countries.

The purpose of my project

==> Why is illegal logging a problem ?
==> Between 25% and 30% of the greenhouse gas released into the environment come from deforestation, More than one in four people around the world relies on forest resources for their livelihoods. Yet forests have effectively disappeared in 25 countries and another 29 countries have lost more than 90% of their forest cover.

==> The main purpose of this project is to use the current technology and save the forests around the globe.

How it works

As the forests are generally of large areas, it's difficult for the rangers to safeguard the whole forest. The main aim is to alert the forest rangers if any vehicle, chainsaw or some other machinery sounds were heard, so that the rangers can quickly go to that particular location and catch the illegal loggers. It is difficult for a human being to keenly listen to the surrounding sounds 24/7, to make life easy, A machine learning model is build, which is capable of classifying the surrounding sounds, using this model and some other sensors we can alert the officials/ forest rangers if any vehicle, chainsaw sounds were heard, We need to fix the modules(Arduino/Raspberry pi/Recycled phones)on the trees, which has a microphone, solar panel/batteries, gsm, gps module and this machine learning model loaded. Each module has a range of 500m to 1Km, so we can fix the modules accordingly.

The languages I used

I build it using python. I used machine learning architecture of A multilayer perceptron (MLP) ,a class of feedforward artificial neural network (ANN) and a convolutional neural network (CNN, or ConvNet), a class of deep neural networks while also making use of prominent Python Libraries such as Tensorflow, Keras, numpy, Pandas, etc.