Once again, our colleague Thiago Alves presented a session at the ML Conference, but this time invited by the conference team. The event, which takes place three times a year on different cities, was organized in Berlin this time. At the conference, basic and advanced ML techniques are discussed, as well as business strategies based on these technologies. The conference is held over three days where workshops and more than 25 sessions are offered to the attendees.
The session presented by Thiago is the result of his master’s thesis and is entitled Honey Bee Conservation using Deep Learning. During the 45 minutes of the presentation, details of how he and his advisers used Deep Learning to count and classify cells from bee combs were presented. With this project, called DeepBee, beekeepers will be able to spend less time collecting information about the health of their colonies, a task that was previously done manually, and will be able to invest their time investigating ways to uses the collected data to improve their production.
This work was recently published in the magazine Computers and Electronics in Agriculture under the name Automatic detection and classification of honey bee comb cells using deep learning. The source code, models and datasets were made available openly on GitHub.