Based Brothers

https://docs.google.com/presentation/d/1rPwDKII7NiRf57wgHcQrn72RLU-kRJqSYvb882yuBKs/edit?usp=sharing

https://www.youtube.com/watch?v=rXrfGFnfgUs&ab_channel=Apelan

DebriSearcher

The aim of our project is to locate and decrease the level of marine debris by using artificial intelligence and machine learning. In order to comprehend the concept of our project, we need a deep understanding of AI/ML and how do they work.

What is Artificial Intelligence?

Artificial intelligence is the common name of a program that leverages computers to mimic the problem-solving and decision-making capabilities of human minds [1]. AI is not as perfect as human minds but that makes it more useful and easy to use since it can be used for particular purposes.

What is Machine Learning?

Machine learning is an area of artificial intelligence and computer science that focuses on using data and algorithms to mimic the way humans learn, with the goal of steadily improving accuracy [2]. Machine learning is similar to our brain and neural system. It was developed by Arthur Samuel in 1959 [3].

How Does Neural Systems in Machine Learning Work?

Neural network in machine learning is not different from the human neural system. There are at least 3 columns of neurons with similar but different functions. Those columns are named input, processor, and output neurons. Input and output neurons are called visible layers which is necessary for all neural networks. Processor neurons are called hidden layers and can have more columns than other layers. The number of layers and neurons are the main factors for efficiency in machine learning. Next image is an example of neural network in machine learning [4].

After preparing AI and ML bot for our app, we will use the bot to find the locations of polluted areas visually. The data that we get from the bot will contain 5 pieces of information.

Depth of The Sea/Ocean

Knowledge about sea level and depth are important for our bot to predict the best solution. For our purposes, our bot will use a visual prediction method for undersea level and depth. Normally, according to NASA's statement, satellites are used in the calculation of depth but our bot will estimate the level by its color tone. Next image shows how satellites measure the depth of a sea [5].

The Density of Plastic Pollution

The plastic density of a polluted area is also an important factor for our AI. Density is a necessity for our bot to choose between governmental and communal solutions. Dense pollutions might not be solveable with communal solutions.

Distance Between Closest Seashore and Polluted Area

The horizontal distance between polluted area and closest coast is important for better predictions. The recommended solution might change according to distance. If polluted areas are distant, government support for shipment might be a necessity.

Diameter of Polluted Area

Finding the diameter of the polluted area is one of the most important factors in the prediction process. Larger numbers might need more governmental solutions.

Average Size of The Plastics

The average size and diameter of plastics are also important for calculation. We can assume that the average size for communal solutions might be 5-10cm, other numbers may require government help for solutions.

Programming Languages

For our purposes, we decided to used different languages and modules for different requirements. Application frontend code, we will use Javascript, React and React Native. Data and backend is going to be handled with Node.js and MongoDB. For AI/ML, we might use Javascript or Python. Since these languages are incompatible we will try both of them and decide on which one is better for us.

How does our project solve the problem?

Each step to the solution:

• The visual AI bot will use a camera to search for polluted areas. For that camera, we might use drone cameras to speed up the process.

• Located region data will be stored according to different criteria: Coordinates of the located area, depth of the ocean, the density of plastic pollution, distance between closest seashore and polluted area, diameter of polluted area, average size and detection date.

• Obtained data will be used inside the app to create different types of pins. Those pins will show danger levels. Coordinates and date will be used for backend process.

• With danger levels, our AI will predict and choose the best possible solution way and we will present it in our app. Also our app will contain information about possible harms that may occur.

Vulnerabilities

There are some problems our app will not be able to solve. Our project can not solve the problem for microplastics since they are almost impossible to detect visually. In our future versions, we might try to solve this problem with optical light refraction.

Project Introduction Video

https://www.youtube.com/watch?v=rXrfGFnfgUs&ab_channel=Apelan

Project Code

https://github.com/Kungie/DebriSearcher

Data and Methods From NASA

While thinking about our project, the websites NASA provided inspired us and let us have better visions about the topic. Our thoughts about marine debris and plastic pollution changed and we started to understand the importance of purification for our seas. The data we will use in our app will be our own data but some of the methods such as calculations about depth, diameter and density are from websites.

Our dedication to saving, cleaning oceans, and the world has never been a temporary enthusiasm it genuinely persisted since any of us had enough capability to sense conflicts -contradictions- in imminent environments, therefore, encountering this challenge was a perfect fit and made us feel fulfilled simultaneously. Eventually having the opportunity to talk and propose our opinions as much as we like was like a dream, times when we just stood there and fear no one was listening has ended because we had each other, every member of our team was ready to listen without any prejudice. Despite having some ideas being given a limited time to sort many problems and apply these solutions into a perceptible realm requires us mentees to act in a hustle thus we enhanced our ways and the duration of these paths in working with harmony, consistency. Of course, there were interferences and obstacles during the progress however deriving new proposals, compromising in different areas is what made our team unique. In conclusion, everyone who contributed to this project will hopefully continue living with curiosity and someday prove their virtuosity until then this ginormous experience is going to be neither a beginning nor an end. It shall stay as the tip of an iceberg beside our future developments.

We'd like to thank TED Çorum College for generously informing us, TED Mersin College for their support and NASA for giving us the opportunity to compete in Space Apps Challenge.

URLs provided by NASA

https://www.nasa.gov/feature/esnt2021/scientists-use-nasa-satellite-data-to-track-ocean-microplastics-from-space

https://www.doi.gov/ocl/marine-debris-impacts

https://coast.noaa.gov/states/fast-facts/marine-debris.html

[5]. NASA sea level calculation and image - https://earthdata.nasa.gov/learn/articles/pod-sea-level-rise

Other URLs:

[1]. What is Artificial Intelligence - https://www.ibm.com/cloud/learn/what-is-artificial-intelligence

[2] What is Machine Learning - https://www.ibm.com/cloud/learn/machine-learning

[3]. Who is ML's Developer - https://www.cognizantsoftvision.com/blog/what-is-machine-learning/

[4]. Neural Network Image - https://www.researchgate.net/figure/An-example-of-a-deep-learning-neural-network-with-3-hidden-layers-For-a-Boltzmann_fig6_286513346

#debrisearcher #ai #ml #marinedebris #plastic #pollution

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