Inspiration
Marine debris, also called marine trash, is any human-made solid material that is disposed of or abandoned on beaches, in waterways that lead to the ocean, or in the ocean itself, regardless of whether disposal occurred directly, indirectly, intentionally or unintentionally source.So we decided to do something about this.The best way was to encourage people in disposing off wastes thrown at beaches.So we created a website based on this.
What it does
Captain Blahaj's is our saviour superhero. Capt's website asks its users to upload a photo of the waste which you'll be disposing that you found at a beach.The site's backend then analyses the image using Microsoft Cognitive Services API to verify whether the image contains a beach/waste and accordingly reward users with points. This will encourage more n more people to dispose waste found at beaches and help protect the ocean and marine life.The site also features a leader board of all users with the data stored on Datastax's cloud db. You can also share your achievement on twitter so that others also notice your achievement
How we built it
We built the front end using HTML,CSS,Js . The backend API was made using flask.The Image uploaded is sent to the API via a POST request and the backend uses Microsoft Cognitive Services API to obtain the labels in the uploaded image.If the image contains a beach/waste etc then the user is rewarded with Blahaj points for doing their part in Captain Blahaj's Mission to save the seas.
Challenges we ran into
- Since our hack involved analyzing the image sent and none in our team are familiar with machine learning.We decided to use google cloud vision API but we didn't have a credit card to avail it.Then came Azure's student account to the rescue!!.We used Cognitive Services API and voila it is as simple as sending a POST request.
- We needed to store our users score in a db.We weren't quite familiar with using databases in the back end.Then we decided to give Datastax a try and their docs being much beginner friendly helped us achieved our requirement.
- Updating HTML based on the API response was a bit tedious task for us.
Accomplishments that we're proud of
- We are proud that we could finally achieve of what we had planned to achieve!
- Successfully used a backend db reading the documentation/ (:D).
- Created an aesthetically appealing front end
What we learned
- How to send POST request from a form element
- What is CORS and why it blocks Cross domain requests.
- How Machine Learning works and what are labels.
- How to use datastax cloud REST API.
What's next for Captain Blahaj
- Make a fully functional User panel for the User with much more tasks.
- Update the front end after submitting the file instead of an Alert.
- Make use of datastax's cassandra python driver instead of using the rest api.
- Create a Custom ML model that can detect wastes thrown at beach like plastics,bottles etc.
Built With
- azure
- css3
- datastax
- flask
- html5
- javascript
Log in or sign up for Devpost to join the conversation.