-
-
one of the app's amazing creators showing off the camera feature
Inspiration
We wanted to use AI to educate people on the sustainability of the environment by informing them about the state of plant life around them. By combining that concept with the benefits of experimental learning, we figured this would not only engage students in the field but also make environmental concepts exciting. This goes into specifically Agricultural/Environmental sciences education curriculums, also can be used for younger students in environmental subjects.
What it does
You take a picture of an environmental terrain/plant and the underlying ML model predicts the health index of the plant/plants detected in the image. Then an AI agent (if it has been given a user fed image of the plant/terrain) provides additional info such as possible reasons for the health index, and answers any of the user's questions.
How we built it
using html-css/javascript for visuals and then python libraries to access LLM models for plant analysis and the AI assistant
Challenges we ran into
putting together the base cnn model with the javascript visuals was difficult.
Accomplishments that we're proud of
Creating a complex model
What we learned
more coordination and preperations saves a lot of time.
What's next for Immersion
Expanding beyond vegetation health and more environmental analysis including drought percentage, wildfire chances, etc.
Built With
- css
- googleai
- html
- javascript
- pil
- python
- pytorch
Log in or sign up for Devpost to join the conversation.