AI-Binovator

Inspiration

The idea comes from the growing environmental problem which many people are lack of awareness or education for throwing the a recycle rubbish to the sustainability garbage specifically to the elder and children. Thus, the AI sustainability garbage will help in auto recycle the rubbish when people throw it to the garbage. Furthermore, mobile app for collecting points is developed because the activity of points earned after throwing the rubbish will encourage more people use the developed garbage which helps in decreasing the population. When the points earned until the certain amount, points can used for exchange prices.

What it does

1) Take the rubbish photos 2) Analysis the type of rubbish. 3) Send points to app.

How we built it

We differentiate into 3 parts which each of the members responsible in one task:

• Image capture on the garbage device - A smart bin that capture the image of rubbish
• AI Model and Classification - A machine learning model trained to analyze the captured images and classify the waste into categories such as paper, plastic , glass, metal, organic, hazardous.
• Mobile App Development - A user-friendly app that connects to the bin, tracks user points, and allows point redemption.

Challenges we ran into

The challenges we faced was establishing smooth communication between the hardware, AI model, and mobile app specifically integrating Kiro's part with the rest of the system. This is because we had not worked together on developing apps or AI models before so we faced a steep learning curve. Therefore, we spent a lot of time in learning, testing, and troubleshooting during development.

Accomplishments that we're proud of

Successful developed an almost completely AI model which we learning at the first time and this is the first project we made by our own.

What we learned

• the integration and use of API between each application
• Code to connect with the camera device for image capturing
• Team collaboration and time management

What's next for -

• improve the accuracy of the AI model as we can't completely trust the AI agent as may contain biases which unable completely handles until now
• include the QR code for connecting the mobile app which we unable to complete in our project
• include the voice assistance and multilingual support to help elderly users interact with the bin.

Built With

• html
• javascript
• kiro
• kotlin
• python
• xml