Harvest Vision is a machine learning and computer vision approach to make harvesting more productive.

What it does

HV provides real-time feedback to mechanical harvest drivers to boost yields which translate into additional revenues

How I built it

This project was built using a Raspberry Pi, a Neuromem chip, and Aloha. Our technology stack included Python, OpenCV, HTML, CSS, and Javascript. We used open-source libraries to enhance our UI such as Chart.js, D3.js, and Request.

Challenges I ran into

Perfecting the training data and handling REST APIs to process display data. One of the difficulties was edge detection and color classification.

Accomplishments that I'm proud of

Creating a project using machine learning to solve a real-world problem within 24 hours.

What I learned

Anything is possible with teamwork.

What's next for Harvest Vision

Kauai Coffee the largest coffee producer in the USA will pilot Harvest Vision in August 2018. Another application is a coffee bean inspection system for the Dept of Agriculture.

Built With

Share this project: