Awards & Nominations
Yellow Submarine has received the following awards and nominations. Way to go!
Global Nominee
Yellow Submarine
High-Level Project Summary
We developed a project that collects plastic among the waste products. As we have found, there is an average of 12% of plastic in total in every litter found. At least 800 species worldwide are affected by marine debris, and as much as 80 percent of that litter is plastic. Our project solves this problem via identifying unique and specific objects like plastic bags and bottles. It also utilizes the potential advantages of AI and ML to collect plastic which is an efficient way of identifying objects. Our project uses YOLO4 Tensorflow in object detection which recognizes objects more precisely and faster than other recognition systems. It can predict up to 9000 classes and even unseen classes.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
With the usage of 300 photos, the project is trained for approximately 6 hours. After this time period, the robot can identify the trained object precisely. Due to the fact that our personal computers do not have sufficient power, we trained them via google collab. Our main aim is to create a system of robots that will identify on the map which regions are mostly polluted. In addition to this, our system can precisely identify what is plastic and what is not. For developing this project we used python and jupyter for data analysis, NASA data sources and several other platforms.
According to the United Nations, at least 800 species worldwide are affected by marine debris, and as much as 80 percent of that litter is plastic. It is estimated that up to 13 million metric tons of plastic end up in the ocean each year—the equivalent of a rubbish or garbage truck load’s worth every minute. Fish, seabirds, sea turtles, and marine mammals can become entangled in or ingest plastic debris, causing suffocation, starvation, and drowning. Humans are not immune to this threat: While plastics are estimated to take up to hundreds of years to fully decompose, some of them break down much quicker into tiny particles, which in turn end up in the seafood we eat. This is the reason why we chose to battle with this challenge and we want to see the change in our world through our systems.
Space Agency Data
NASA Data for mss: https://podaac.jpl.nasa.gov/CYGNSS?tab=documentation§ions=about%2Bdata,
specifically : https://podaac-opendap.jpl.nasa.gov/opendap/allData/cygnss/L2/v2.1/2021/contents.html
NOAA GDAS for wind_speed: https://rda.ucar.edu/datasets/ds083.3/#!description
On what this algorithm is based: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9449485
Hackathon Journey
The experience was awesome! We learned not only new programming skills but also teamwork and leadership skills. Our team found balance in every aspect of the project and consists of universally great people from Kazakhstan who have the same aspirations and goals:) Our main approach was not only implicating what we know but also learning something new from others. We would like to thank American Corner for organizing such an event! Being driven to battle climate change through developing an efficient system helped us to unite and become a team.
References
1) For web: React JS, jsx, json, Microsoft Studio. For data analysis: python, jupyter
2) NASA Data for mss: https://podaac.jpl.nasa.gov/CYGNSS?tab=documentation§ions=about%2Bdata,
specifically: https://podaac-opendap.jpl.nasa.gov/opendap/allData/cygnss/L2/v2.1/2021/contents.html
NOAA GDAS for wind_speed: https://rda.ucar.edu/datasets/ds083.3/#!description
On what this algorithm is based: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9449485
Global Judging
This project has been submitted for consideration during the Judging process.