Trash-Snap

Inspiration

Every year, a staggering amount of waste is improperly disposed of, leading to environmental and economic consequences. Did you know that globally, E-waste comprises a staggering 70% of our overall toxic waste? Shockingly, only 12.5% of E-waste is recycled, while a staggering 85% ends up in landfills or is burned, releasing harmful toxins into the air. The consequences are dire, as electronics contain lead, a hazardous substance that can damage our central nervous system and kidneys. For instance, a study by CBC News revealed that in Canada alone, up to 30% of recycling placed in blue bins is contaminated with non-recyclable materials. Furthermore, people often dispose plastic waste incorrectly. According to a report by Our World in Data, of the 359 million metric tons of plastic produced in 2018, an estimated 91% was not recycled and ended up in landfills or the environment.

These alarming statistics motivated us to take action and develop a solution that would help individuals identify the specific type of waste they have and help them dispose of it. Our solution, Trash-Snap, is an application aimed at revolutionizing waste disposal practices and reducing environmental contamination caused by hazardous materials.

Trash-Snap is more than just a convenient accessibility pro application; it's a powerful tool for change. Our mission is to provide a platform that enables individuals, regardless of their technological proficiency or geographic location, to easily and correctly dispose of their waste, including e-waste. By doing so, we can collectively minimize the harm caused by toxic materials and promote a more sustainable future.

But our vision doesn't stop there. Trash-Snap is just the beginning. We have a plethora of exciting features planned for future iterations of the application, ensuring that we level the playing field for everyone. We want to ensure that every individual has the opportunity to participate actively in creating a better, more sustainable world.

Trash-Snap is an invitation for collaboration, an opportunity for us all to unite and work towards a common goal. Together, we can forge a path towards a more equitable and sustainable future, not only for ourselves but also for future generations. We sincerely hope that our project serves as an inspiration for others to take action and make a positive impact in their own communities.

Let's harness the power of technology and collective action to create a better future for ourselves and our planet. Join us in this transformative journey and let's make a lasting change. Together, we can build a world where waste is managed responsibly, environmental contamination is reduced, and the well-being of both people and the planet is prioritized.

What it does

The webapp that allows users to conveniently scan their garbage, recycling and compost items OR bins and be directed to the correct way to dispose of the item. We were to able to implement it for hazardous waste such as e-waste due to time constraints, but we hope to do so in the near future. When scanning bins, not only is the type of bin identified. We have an added feature of a chatbot to accommodate user preferences as well as use cases where one may not want to open their camera.

How we built it

-created apis from django(backend)

-ajax tech is used so as to prevent from whole page reload every time we need some data -boostraping for quick ui model

-hugging face >> classification model . we used api interface to classify the image instead of using whole model in local

-opencv

folium from python external module helps to build map (open street map -open source)

-voiceflow for user interaction for any faqs

Challenges we ran into

Implementing our ideas came with its fair share of challenges. With remote team members spanning different time zones, coordinating became a juggling act. Moreover, our ambitious vision proved to be a test for our first-time hackers, as they faced unfamiliar tasks. Nevertheless, we stand proud of our accomplishments. It was hard to integrate both the backend and the frontend. None of us had experience using ajax to send requests to the django backend.

Accomplishments that we're proud of

Our team had an amazing time at this Hackathon, and we're thrilled to share our accomplishments with you! Despite being complete strangers, we quickly bonded over our love of technology and desire to make a positive impact.

One of our proudest accomplishments was working together as a team to tackle each challenge head-on. We were able to seamlessly delegate tasks and use our unique skill sets to make this project come to life. We also had a blast learning about all of the new technologies we used throughout the Hackathon. From object detection to advanced coding techniques, we were able to expand our knowledge and improve our skills in so many ways. The real cherry on top was finally nailing the object detection implementation! After hours of hard work and collaboration, we were able to overcome this obstacle and add the core functionality of our idea. Each member of our team has something unique to be proud of, whether it's creating stunning visuals or crafting expert-level code. We're honored to have been a part of this amazing event, and we can't wait to see where our skills take us next!

What we learned

During the process of creating this project, we learned so much about the importance of taking action to better our planet. We were inspired to see the impact that even small changes can have on the environment, and it reminded us that we all have the power to make a difference. By increasing awareness and encouraging everyday people to be just a little bit more conscious of their actions, we can all take steps towards a brighter, more sustainable future. It was truly inspiring to see everyone coming together at this Hackathon to work towards a common goal, and we feel proud to have been a part of it. We learned that every effort counts, and that together, we can make a positive impact on the world around us.

Here is a list:

Incorporate Figma design
Improve error formatting
Provide a list of instructions/ resources to the user after scanned image is analyzed as well as in the chat
Currently errors present in taking a picture button and clicking the upload button without first selecting an image
Dynamically generate relevant locations on map
Add a feature where AI generated recommendations will show up for the user, model must be trained first
Improve object detection process
Adding more endpoints/ paths to the webpage
Adding users and accounts
Chatbot
Adding in voice capture and speaking capabilities for accessibility
Adding AI functionality
Expanding intent training - currently only trained by 1 dev
Add actions and expand conversation length to suit certain users