Nature Explorer

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

I built Nature Explorer, a Streamlit app that encourages people to step outside, find something in nature, and learn about it using AI.

The idea is simple: take a photo of something you find outdoors, such as a plant, upload it to the app, and let the AI analyze the image.

Nature Explorer provides:

• What the AI thinks the object is
• A confidence estimate
• What it can see in the image
• Interesting facts
• An outdoor discovery challenge

The goal is to turn a screen into a starting point for real-world exploration instead of keeping people on the screen.

Code

https://github.com/bandarumanas/NatureExplorer

How I Built It

I built Nature Explorer with Python and Streamlit, with Gemma as the AI model at the core of the application.

The app allows users to upload an image and sends the image to the Gemma model for analysis. The response is then displayed in a simple Streamlit interface.

The main technologies I used are:

• Python
• Streamlit
• Google GenAI SDK
• Gemma
• Pillow
• python-dotenv

I used the Gemma 4 26B A4B IT model for image analysis.

I also kept the API key outside the source code using an environment file and added it to .gitignore so the secret is not committed to GitHub.

Why Does Open Innovation Matter?

Open innovation matters because it makes it possible for developers to build AI-powered experiences without having to build a model from scratch.

For Nature Explorer, using an open-weight Gemma model made AI a practical part of a small project that encourages people to explore the world around them.

It also gives developers more flexibility to experiment, learn how AI systems work, and build applications around open models.

Best Use of Gemma

Nature Explorer uses Gemma as the core AI model for analyzing uploaded nature images and generating explanations, facts, and outdoor discovery challenges.