MOSKEEL(OddCode)

Inspiration

What’s the most dangerous creature on earth? Without question the answer is: the mosquito. Mosquitoes and the diseases they spread have been responsible for killing more people than all the wars in history. Even today, mosquitoes transmitting malaria kill 2 million to 3 million people and infect another 200 million or more every year. Moskeel is here to fix it.

What it does

We develop a tool that uses machine learning to disperse a spray to kill mosquitoes by listening for the frequency at which their wings flap. Each of our gadgets transmits the data it has gathered to our dashboard so we can keep track of their status and manage them. This information may be utilized to do research, develop solutions to the issue, and identify where the insects are coming from.

How we built it

We trained a tensorflow-Keras convolution neural network with data from HubBugDB which contains a labeled collection of mosquitoes wing sound.

Challenges we ran into

Mosquitoes wings sounds Datasets are hard to find and some of them are not open source.

Accomplishments that we're proud of

We are proud of the model we built, it can detects mosquito sounds even in the midst of noise,

What we learned

We learned about the general things about IOT, and how to use Arduino.
We learned how to use blender and how make simple 3D animation
We improved our skill set in many technologies including React, Tensorflow and we learned some deep learning techniques using constitutional neural networks.