High-Level Project Summary
Using citizen scientist and NASA satellite data to detect and classify marine plastics using TensorFlow. Then visualize the data in interactive charts and create heat maps to aid policy makers and educate the general public.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
The project focuses on the issue of ocean plastic contamination. Tools of AI/ML technology will be utilized to first leverage remote sensing data from NASA's database as a manner of assisting in the reduction of marine debris. The algorithm's job then becomes to find plastics in the data. Information will be collected and sorted into groups based on their sizes, types, and shapes using the same technology. The analyzed data will next be represented in the form of charts and maps. Working on this project will provide vital advantages to both humans and marine species, including safer shorelines, economic growth, increased fish quantities, improved air quality, and a higher quality of life for coastal communities. Finally, reduce, if not eliminate, the suffering of marine life.
Space Agency Data
Data from Sentinel-2 satellites.
Hackathon Journey
We were ecstatic to participate in the NASA space apps Hackathon for the first time. The team decided on the LEVERAGING AI/ML FOR PLASTIC MARINE DEBRIS challenge because of its significant local and global impact. We began working on the project by exchanging ideas with one another and consulting the website's available resources. Our plan was to leverage NASA data on plastics in the water to detect them using AI/ML technology. The data will subsequently be processed and visualized.
However, we ran across several issues during the procedure. NASA data, for example, was difficult to come by. Furthermore, we received no response when we attempted to contact the moderators. As a result, we were unable to complete the project on time.
Additionally, we weren't able to get connected with our local leads as we were only wait-listed. Which lead to our confusion and having to take multiple approaches and start over multiple times. All said, it was still an enjoyable learning experience and we wish to have a better outcome in the next event.
References
https://www.nature.com/articles/s41598-020-62298-z
https://github.com/antiplasti/Plastic-Detection-Model
Tags
#AI #ML #tensortflow #geosensing #visualization
Global Judging
This project has been submitted for consideration during the Judging process.