#NonStopHer: “To solve a global problem and make an impact, we need to push ourselves”, say Mihika Shrivastava and Kaaviya Ramesh, 3rd-year Computer Science students.
Hackathons are something which, as engineering students, we’ve always known about but we’d never participated in. So, when our friend (and teammate) Swetha suggested that we take part in this, all of us- Mihika, Gomathi, and Kaaviya, were immediately on board.
Hackathons always seemed daunting, yet exciting, with the expectation to create something meaningful that could solve a global problem and make an impact all within a limited time frame, pushing you to dive headfirst into the problem and its potential solutions- Yes, please.
The general topic for this hackathon was Sustainability, Focusing on the UN’s 17 sustainable goals. So as a team, we spent quite a lot of time debating and discussing the goals and problems we could choose. We ended up choosing Goal no. 11- “Sustainable Cities and Communities’, under which the issue we took on was uncollected and untreated garbage.
And the statistics support it. India generates about 62 million tonnes of municipal solid waste annually. Only 70% is collected.
Despite living in 4 different cities, we all had faced one common problem: the sight of garbage dumps. And the statistics support it. India generates about 62 million tonnes of municipal solid waste annually. Only 70% is collected. Only 11.9 million tonnes are treated. The remaining 31 million tonnes are dumped in landfill sites. And the uncollected 30%, which is an astounding 18.6 million tonnes, is found on street sides.
So to tackle this problem, we had to answer two questions essentially:
Q1. Where is the garbage?
As India and the world urbanizes, there is a sudden rise in the number of corporations providing deliveries to people in and around cities during the pandemic. This puts us in a unique situation wherein we can, with the help of the corporates that provide the delivery, observe locations wherein unnoticed garbage is collected.
Q2. How would we identify the garbage?
To identify the garbage, we decided to use a Deep Learning model. Our set-up to place said waste would be fitted on the sides of delivery vehicles. It would comprise a smart camera to see the streets, GPS to detect the location, and a modem to send the area to a Database server so that the coordinates are noted and displayed on a map for the convenience of the concerned authorities to clean up. These authorities could be municipalities or recycling plants.
we decided to implement a small-scale version of our idea by using the YOLOv5 algorithm, which we trained using a self-created data set
So once having answered the above two questions, we decided to implement a small-scale version of our idea by using the YOLOv5 algorithm, which we trained using a self-created data set (using IBM’s cloud annotation software) to recognize six different types of garbage- Glass, Metal, Plastic, Cardboard, Paper and Trash. We used IBM’s database to store the locations and python to show the spots on a map.
With this solution in place, fingers crossed, we submitted our idea for Round 1. So after about a week, we heard back saying we were selected for round 2, and we were beyond excited. We started to work further on our idea, considering possible future expansions and possibilities which could arise with the onset of 5G. After our presentation for round 2 was done, we heard back the same evening to give our photos which got us even more pumped for the future of this idea.
We learned that we were selected in the top 12 teams out of over 7000 groups across the country.
With each hackathon round, we received constructive criticism and encouragement, which pushed us further to improve our idea and do our best. We are grateful to have gotten this opportunity, and we would like to thank WiT-ACE for it.
Authors: Mihika Shrivastava and Kaaviya Ramesh (We, Mihika Shrivastava and Kaaviya Ramesh are 3rd-year Computer Science students at Amrita Vishwa Vidyapeetham University, Coimbatore. We are both computer science enthusiasts who enjoy reading and exploring new avenues, academically and artistically.)
Connect with us on: Facebook, LinkedIn, Twitter, and Instagram.
