Green Excel-ence

I worked on a project called Green Excel-ence, where I trained a machine learning model on my laptop to recognize specific visual patterns — in this case, telling whether a fruit is fresh or rotten. Once the model was ready, I deployed it onto a MAX78000 microcontroller, a small but powerful device with a built-in camera and button. Connected to my laptop via PuTTY, the microcontroller captures real-time images of fruits, processes them locally using the trained model, and sends the results back. While it processes data on-device, it still relies on the laptop connection for communication and overall operation.

This setup demonstrates the potential of edge AI, even when paired with a supporting system. It enables faster analysis, reduces the need for heavy cloud processing, and offers greater flexibility for testing and deployment. Such technology could benefit farmers, food suppliers, supermarkets, and even households, helping to reduce food waste, maintain quality control, and ensure fresher produce reaches consumers. Green Excel-ence is a step toward making AI-driven quality checks more efficient and accessible.