Mines are some of the most attractive weapons available to any determined adversary and represent one of the most vexing military challenges.
Sea mines are perhaps the most lethal form of these weapons, as they are hard to find, difficult to neutralize, and can present a deadly hazard to any vessel.
So here we come in with DodgeMine to predict the best possible route while travelling in a marine vehicle.
💻 What it does
The machine learning system, which is based on the neural network, needs a training data set that consists of samples of rocks and mines.
The training set, which has a distance between the samples, is given to the system, which results in a classification system where each sample has a score that shows its closeness to the other samples.
Based on this closeness, the system predicts a safe pathway between the two samples. The system, based on the machine learning model, consists of two parts: one for analysing the sonar radiation and the other for classifying the samples.
This system can be easily applied to a variety of applications that require intelligent classification.
⚙️ How we built it
The Client end of our website was built by using HTML,CSS and JS.
We used Logistic Regression for training the model.
We used Flask for our server to integrate an ML model with our website.
🧠 Challenges we ran into
Getting datasets for detecting mines or rocks was a bit difficult.
Selecting a perfect machine learning algorithm was a hard task.
As we are new to flask, We spend more time in integrating our model with our front end.
📖 What we learned
We learned to train a model within a short span of time.
We learned flask how to integrate a ML model with the client end.
We learned about the pickle package in Python, which is used for prediction.
📧 Use of Twilio
We used Twilio to send mine vs rock report to our users.
Twilio is safe and secure API for sending text messages.
☁️ Use of Google Cloud
Google Cloud offers Machine Learning and Deep Learning models.
We used google cloud logistic regression machine learning model to train our model.
📖 Use of Deso
Deso is a decentralized social application and it is open source & on chain open data
We used deso for login, logout purpose and also for transactions occurs in our website.
🚀 What's next for Dodge Mine
To upload input data as a file format.
To feed the model with more datasets and to increase its accuracy.
To test the dataset with different algorithms and to find the optimal algorithm.
🏅 Accomplishments that we're proud of
We're glad to sucessfully complete this project!
The end goal was met to a satisfactory level, and the outcome would allow help seaman to detect mine accurately and
make them to have a safe jorney.
Log in or sign up for Devpost to join the conversation.