Skip to main content

Using Artificial Intelligence for Strawberry Yield Prediction

The University of Florida and NVIDIA have recently announced a transformational partnership to create the most powerful artificial intelligence (AI) supercomputer in higher education. UF Agricultural and Biological Engineering (ABE)’s work in AI research, like all research in ABE, seeks to transcend disciplinary boundaries to create synergy among different knowledge areas for designing, quantifying, assessing, and managing engineering solutions for natural and managed systems.

To spotlight the research being done, our ABE Blog will be featuring a series all about our work in AI and the faculty who are striving for excellence in this field. First, we are featuring ABE Professor Won Suk “Daniel” Lee and his work on strawberry yield production.

Strawberry Yield Prediction Models Based on Imagery Information

The overall objective of this project is to improve strawberry yield prediction models using field images and other variables. Using computer vision and artificial intelligence (Faster R-CNN model), strawberry flowers and fruit are automatically detected from images acquired from the field and their numbers are counted. These numbers are used in yield prediction models.

During a growing season, growers need to know when major fruit waves (i.e., a lot of fruit to harvest) come so that they can prepare proper harvesting labor and plan for marketing. One way to find out is to count the number of flowers at a certain time and they can expect near-future yield using yield prediction models. The newly developed AI-based method replaces manual counting of flowers and fruit with automatic and faster counting.

Imaging cart featuring GPS receiver, computer, battery and inverter, camera, and lights.

The ultimate goal of this AI research is to develop a robotic system that can autonomously navigate strawberry fields, take images, create a distribution map of flower and fruit, and predict yield. The system will be developed in a similar way that growers operate it using a smartphone app.

Within this research, students are being trained to understand how the AI algorithms work and learn how to write a code for necessary operations and apply different AI methods for various agricultural operations. The recently announced world’s most advanced AI system at UF will play a crucial role in training students.

Strawberry Detection

This project is funded by the Florida Strawberry Research and Education Foundation. This team includes Dr. Won Suk “Daniel” Lee, Feng Wu, Amr Abd-Elrahman, Natalia Peres, and Shinsuke Agehara.