High-Level Project Summary
What did you develop? We developed a project that consists of a Neural Network that gives us the benefits of computer vision, which allows us to quantify the accumulations of garbage as well as to know their location by geolocatable coordinates at least the prototype.How does it "solve" the challenge? Knowing the amount and location of these garbage piles we can collect this garbage using techniques already implemented in other countries such as "Ocean Cleanup System 001". Once we have this garbage, we can compact them.Why is it important? Marine plastic pollution causes severe damage to marine life causing the suffocation of different marine creatures and also the health of human beings.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
What exactly does it do?
We developed a project that consists of a Neural Network that, unlike the others that have been developed, offers us the benefits of computer vision, which allows us to quantify the accumulations of garbage as well as to know their location by geolocatable coordinates. minus the prototype. We think that through photogrammetry development we can calculate the volume of the waste for a more precise quantification.
Currently our program allows us to know the environmental conditions of a specific point to know if it is convenient or not to let the drone patrol the coast since the strong wind and other adverse conditions can cause it to be lost in the ocean.
It should be clarified that we plan in the final version to be able to foresee the future position of the garbage through machine learning techniques, feeding the network with the environmental conditions to be able to predict the future position of this waste.
How does it work?
First, a Neural Network detects the accumulation of garbage from the detection of objects, then computer vision calculates area but later we will calculate the volume, and thus we will quantify the amount of garbage and its location.
Knowing the amount and location of these garbage piles we can collect this garbage using techniques already implemented in other countries such as "Ocean Cleanup System 001"
Once we have this garbage, we can compact them into blocks for sale to the companies in charge of recycling.
What do you hope to achieve?
STAGE 1: We are looking for organizations such as the Ministry of Innovation, Ministry of the Environment, MINEDUCYT, INJUVE and UNES, that allow us high school and university students to receive their social hours by collaborating in picking up garbage at the beginning in beaches and easy areas access.
STAGE 2: Once we have this garbage we can compact them into blocks for sale to the companies in charge of recycling it.
STAGE 3: With the capital raised by the sale of these blocks to recycling companies, we will be able to improve the process to deploy drones to know the location of these plastic waste in the ocean and thus collect the islets by proximity to the coast and quantity, and with Our own plastic crushing and centrifugal machines that we were able to obtain from the capital raised, we will be able to gather we will be able to sell the pellets to Salvadoran companies, making the recycling process more profitable.
What tools, coding languages, hardware or software did you use to develop your project?
All the code was programmed in Python language and the tests were executed on the Google colab servers, the libraries used are the following:
cv2
numpy
google.colab.patches.cv2_imshow
geocoder
Space Agency Data
We use data from the publication "Notes from the Field", "Plastic Ocean", to verify where the promoters of marine plastic debris are in the world and to verify the amount of existing plastic.
https://earthobservatory.nasa.gov/blogs/fromthefield/2012/09/18/plastic-ocean/
Hackathon Journey
How would you describe your experience with Space Apps?
It seemed like an opportunity for development and to make our ideas known
You learned?
Much about the projects that are currently being implemented and about programming techniques.
What inspired your team to choose this challenge?
The Artificial Intelligence part
What was your approach to developing this project?
Try to use recent technologies to improve detection, apply computer vision techniques.
How did your team resolve setbacks and challenges?
Spreading the work even though we're already late
Is there someone you would like to thank and why?
To our families and friends who supported us.
References
All the code was programmed in Python language and the tests were executed on the Google colab servers, the libraries used are the following:
cv2
numpy
google.colab.patches.cv2_imshow
geocoder
We also saw video tutorials that allowed us to correctly program.
Video tutorials
https://www.youtube.com/watch?v=ql00qM6h7k8&t=1826s
Tags
#hardware, #water, #machinelearning, #artificialintelligence, #ai, #datascience, #python, #deeplearning, #technology, #programming, #pythonprogramming, #programmer, #ml, #developer, #software, #coding, #computerscience
Global Judging
This project has been submitted for consideration during the Judging process.