High-Level Project Summary
The project aims to classify and track marine debris by using ML algorithms and Artificial intelligence, Deep learning and cloud computing.we use algorithms like Random forest , neural networks(python, Jupyter netbook,TensorFlow) in order to classify marine debris and also track it.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
The rapid growth of satellites provide massive amounts of data.Machine learning algorithms and cloud computing help the image processing of satellite remote sensing, and that would help tracking and classify marine debris.
The machine learning algorithms were implemented using TensorFlow and Google Earth Engine platform.
neural network would be also such a good solution to detect and track the marine debris .
unfortunately ,we couldn't obtain food performance due to the dataset size and our limited laptop performance (large time in preprocessing)
Space Agency Data
https://podaac.jpl.nasa.gov/dataset/SMAP_RSS_L3_SSS_SMI_8DAY-RUNNINGMEAN_V4
https://internationalmarinedebrisconference.org/index.php/remote-sensing-of-marine-debris-and-its-drift/
"Next Generation Mapping: Combining Deep Learning, Cloud Computing, and Big Remote Sensing Data."
it was such an amazing article that helps us (they use TensorFlow and Google earth engine platform which helps us to make the approaches of remoting sensing using these tools clear and comprehensive)
Hackathon Journey
Our team is a group of beginners who decided to participate with everything they have in the "Space App Challenge" hackathon, which in turn was a completely new experience to us, but it was such a wonderful and amazing experience for sure we learn a lot of stuffs,especially : remote sensing data and how to deal with it, Geoscience and Remote Sensing, image processing...
Marine debris is one of the most pervasive threats to marine life and humanity, so we decide as a team to choose this challenge due to its importance.
We used Machine learning algorithms and Artificial intelligence, Deep Learning and Cloud Computing in order to classify and detect marine debris from a different dataset; we used Jyputer notebook(python) and TensorFlow.
Even if we were just a group of beginners and newbies, we collaborate with each order in order to reach our goal.
References
Tags
#marine #debris #ML #TensorFlow #Python #googleEarth
Global Judging
This project has been submitted for consideration during the Judging process.