The Green Coders

Cairo | Leveraging AI/ML for plastic marine debris

Saving Nemo

High-Level Project Summary

We are happy to announce that we developed a solution that may help alot in saving marine treasures:1- We Created an AI Model that classifies sea debris to 5 different materials : plastic , metal , paper , cardboard and trash.2-based on this classification , we can determine priority of cleaning this area , send to concerned organization to clean areas with "high priority" example : area that have black opaque plastic bags , that prevents daily sunlight to algae and fishes or area that have many cigarette buds , small particles that fish can eat is considered "high priority"3-We Created a user-friendly mobile application with a detailed geo-spatial map which user can use to track & log.

Link to Project "Demo"

https://prezi.com/view/PiLsuulb4DFQJMRWjYF9/

Link to Final Project

https://github.com/Hazem1234u/Saving-Nemo

Detailed Project Description

1) AI Model:

language : Python

Used Keras , Tensor flow .

Worked on Spyder & Google Colab ( to train model as it has faster GPU acceleration)

function : classifies sea debris to 5 different materials : plastic , metal , paper , cardboard and trash using cam.


2)Mobile Application:

language: Kotlin on Android Studio

Features:

1.Tracking a detailed geo-spatial map for sea debris to participate in cleaning seas . (time-stamp , material , sea depth , cleaning priority) are provided for each pinpoint.

2.Adding debris using mobile phone camera which scans the object to automatically classify material and add it to the map. ( in the future plan)


Project Benefits:


  • Classifying sea debris materials.
  • Identifying priority to clean each area in seas.
  • Having a detailed geo-spatial map with (debris cleaning priority) for each area.
  • Saving Cost & Time by directing cleaning processes in the right direction.



Demo 30-sec Video: https://www.canva.com/design/DAErtf-69zo/kgmQahGfiqrg5Jv44W_rRA/watch?utm_content=DAErtf-69zo&utm_campaign=designshare&utm_medium=link&utm_source=homepage_design_menu

Space Agency Data

  • We Explored NASA Earthdata and we tried the geospatial maps and we were impressed by the amount of useful data like ocean temperature ,winds, waves and chemistry.
  • We Investigated NOAA Marine Debris Program and gained knowledge about garbage patches experiment and project Clearinghouse.
  • NASA Satellite Data technique was a reference tool that provided us with a lot of information to indicate and track the presence of ocean micro-plastic from space.
  • We were inspired by Debris Tracker app and The Global Forest Watch and we managed to explore them in detail to check their features and how it can be improved and how to make an addition to our challenge .

Thank you NASA for providing inspiring and mind opening resources!

Hackathon Journey

This hackathon made us learn that we have no limits , that we have more ability to be productive more than we 've ever imagined . having only 48 hours to implement a solution that may truly help the world was like a challenging mission to serve the world.

We learned that it's not a must to have a complete expertise to start implementing an idea , it's enough to have the passion and the willpower to learn and implement.

My team chose this challenge as we always seen in news and read about sea pollution and how it's affecting sea treasures and how our oxygen may be affected in the future . we are sea lovers with passion to save this beautiful creatures and serve the community with all our knowledge.

Our approach was using YouTube tutorials , calling friends experienced in the work field asking for advices and doing several team meetings to exchange ideas.

References

  1. https://www.nasa.gov/feature/esnt2021/scientists-use-nasa-satellite-data-to-track-ocean-microplastics-from-space/
  2. https://www.nasa.gov/cygnss
  3. https://coast.noaa.gov/states/fast-facts/marine-debris.html
  4. https://sos.noaa.gov/catalog/datasets/marine-debris-garbage-patch-experiment-drifters-and-model/
  5. https://www.nationalgeographic.org/education/programs/debris-tracker/
  6. https://www.globalforestwatch.org/map
  7. https://www.kaggle.com/asdasdasasdas/garbage-classification
  8. https://earthdata.nasa.gov/learn/discipline/ocean
  9. https://colab.research.google.com/
  10. https://www.kaggle.com/kneroma/tacotrashdataset
  11. https://www.youtube.com/watch?v=5Ym-dOS9ssA&list=PLhhyoLH6IjfxVOdVC1P1L5z5azs0XjMsb
  12. https://www.youtube.com/watch?v=ChidCgtd1Lw
  13. https://www.youtube.com/watch?v=VGCHcgmZu24
  14. https://www.youtube.com/watch?v=j_pJmXJwMLA
  15. https://kotlinlang.org/docs/home.html
  16. https://www.tutorialspoint.com/kotlin/index.htm
  17. https://worldview.earthdata.nasa.gov/?v=-141.51146411234396,-48.93945041608877,80.48021411234396,55.40820041608877&l=MODIS_Terra_Chlorophyll_A,Reference_Labels_15m(hidden),Reference_Features_15m(hidden),Coastlines_15m,VIIRS_SNPP_CorrectedReflectance_TrueColor(hidden),MODIS_Aqua_CorrectedReflectance_TrueColor(hidden),MODIS_Terra_CorrectedReflectance_TrueColor&lg=false&t=2017-05-01-T00%3A00%3A00Z

Tags

#nasa_space_app #sea_debris #ai #imageprocessing #artificial_intellegence #classification #object_recognition

Global Judging

This project has been submitted for consideration during the Judging process.

Leveraging AI/ML for plastic marine debris

Marine debris is one of the most pervasive threats to the health of coastal areas, oceans, and waterways. Your challenge is to leverage Artificial Intelligence/Machine Learning to monitor, detect, and quantify plastic pollution and increase our understanding about using these techniques for this purpose.