Inspiration
In South Korea, waste sorting is a very important issue, and from an environmental perspective worldwide, proper waste sorting is crucial
What it does
When the user takes a photo of the recyclable waste they want to dispose of, AI recognizes the waste and provides the appropriate disposal method
How we built it
Frondend : Python flask, java scrip, html, css Backene : Python AI : Pytorch, Yolo
Challenges we ran into
multi waste detection with high accuracy and fast speed. simple and powerful ux/ui
Accomplishments that we're proud of
multi waste detection ux/ui
What we learned
usage of AI at backend server. simple and powerful frontend design.
What's next for Miracle Recycle
increase accuracy increase speed mobile version
Problem Statement: Recycling waste is environmentally important both in South Korea and around the world. Recycling reduces the need for raw materials like timber, minerals, and water. This helps preserve natural ecosystems and prevents over-exploitation of the Earth’s resources. Recycling uses less energy compared to manufacturing with new materials, leading to lower greenhouse gas emissions. This is vital for combating the climate crisis. By diverting waste from landfills, recycling helps reduce methane emissions, a potent greenhouse gas that accelerates climate change. However, properly sorting and disposing of recyclable waste is very complicated and difficult. In South Korea, only about 60% of recyclable waste is actually recycled, and if we were to grade people's knowledge of recycling, it would score only around 50 points. Furthermore, in South Korea, failing to follow recycling regulations can result in a fine of up to approximately 300 dollars.
Team Introduction: Seonghyeon Kim : Backend Developer Juhyeon Lee : Frontend Developer
Project Beneficiaries: It helps people who struggle with sorting recyclable waste to easily find the correct disposal methods for the waste they want to throw away As a result, it can increase recycling rates and contribute to solving environmental issues.
Solution Overview: When people simply take a photo of the waste they want to throw away, AI detects the waste in the photo and provides the correct disposal method. It can detect up to 15 types of recyclable waste, and multiple items can be detected from a single photo
Technical Details: Frondend : Python flask, java scrip, html, css Backene : Python AI : Pytorch, Yolo
multi waste detection, simple and powerful ux/ui
Log in or sign up for Devpost to join the conversation.