Using computer vision to eavesdrop on honey bee communications

Are you interested in the project below? Awesome! PLEASE DO NOT APPLY THROUGH THE URO SITE. Instead, follow this link [] to send in your application by Feb. 20th.

Honey bees are social insects with tens of thousands of bees per colony that have evolved amazing strategies to collectively solve problems. For example, they have a unique communication signal called a waggle dance that allows them to communicate the direction/distance to a rewarding resource. Scientists can "eavesdrop" on these conversations between bees and map the locations that dancing bees advertise. These maps can help us answer questions about honey bee foraging preferences across different landscapes to better understand how we can improve bee nutrition and health. However, manual analysis of videos of waggle dances takes a very long time, limiting the number of experiments we can perform and our sample size per experiment.

To deal with this problem, we have been testing and adapting code that automatically detects and decodes honey bee waggle dances. This Python code was recently developed by researchers at the University of London ( and uses the library OpenCV for computer vision. We have already adapted several processing steps to work with a wider range of videos ( We are currently working to optimize the clustering step. The summer project will involve testing the tracking step, which extracts information about the direction and distance indicated in each dance. It will also involve testing another method for automatically decoding waggle dances that uses OpenPIV, a Python package for Particle Image Velocimetry image analysis.

When we have results showing high accuracy of both waggle dance detection and extracted direction/distance information, we plan to organize/release our version of the code so that other honey bee researchers can easily use it to answer questions about honey bee foraging behavior. We will also also use it to automatically analyze our videos of waggle dances from the Bernard Field Station. We will then combine our dance mapping results with data from another project in the Bee Lab, mapping flowers from aerial drone images, to assess how the distribution of flowers affects honey bee recruitment behavior.

Name of research group, project, or lab
Bee Lab
Why join this research group or lab?

The HMC Bee Lab is an interdisciplinary group that includes students from Biology, CS, Math, Engineering, and other fields studying a wide variety of questions about collective decision-making in both bees and ants. In this lab you will develop new skills and get a sense of many different kinds of research and approaches to answering questions. You could also potentially continue your work in a senior thesis project, present at regional or national conferences, and/or co-author a publication.

Logistics Information:
Project categories
Computer Science
Computer Vision
Student ranks applicable
Student qualifications

All applicants should have a strong interest in understanding the natural world. Second-year students are welcome to apply as well as more advanced Biology and CS students interested in applying their knowledge skills to practical problems. A knowledge of Python programming is required. Experience with OpenCV would be helpful but is not required.

Time commitment
Summer - Full Time
Paid Research
Number of openings
Techniques learned

For this project:

  • Python coding skills
  • computer vision techniques (OpenCV, OpenPIV)
  • data analysis and visualization

All Bee Lab research students will learn to read and discuss scientific literature, and to communicate across disciplinary boundaries and with the public about their work.

Contact Information:
Mentor name
Morgan Carr-Markell
Mentor email
Mentor position
Postdoctoral Fellow
Name of project director or principal investigator
Morgan Carr-Markell, Matina Donaldson-Matasci
Email address of project director or principal investigator
1 sp. | 0 appl.
Hours per week
Summer - Full Time
Project categories
Computer Vision (+2)
BiologyComputer ScienceComputer Vision