Classifying the Queen's Behavior: A Data-Driven and CNN-Based Approach to Analyzing the Honey Bee Queens Oviposition Behavior

This repository contains the Code and Data for the Masterthesis Classifying the Queen’s Behavior: A Data-Driven and CNN-Based Approach to Analyzing the Honey Bee Queens Oviposition Behavior. The results of this thesis were published in Autonomous tracking of honeybee behaviors over long-term periods with cooperating robots. The Repository is split into three parts NeuralNetworkCode: Contains the code for extraction of the candidate events from the compressed data in rosbags. Further the code for candidate annotation as well as the raw annotated data is included. Then the code for Neural Network training as well as the ground truth data in the original split, that was used for training is included. Finally, the code for model application and the full dataset of extracted candidate events that were reported in Autonomous tracking of honeybee behaviors over long-term periods with cooperating robots as well as the trained model that was used for classification is included. PlottingAndDataAnalysis:Contains the code for processing the data and plotting the data for the figures 2.4 and 2.9 as well as the figures A1-A5 in the Appendix. SupplementaryMaterial:Contains selected video files which are referenced in the thesis. License This data and code are under the Creative Commons Attribution 4.0 International license. If you use these data in your work, please cite the relevant thesis Classifying the Queen’s Behavior: A Data-Driven and CNN-Based Approach to Analyzing the Honey Bee Queens Oviposition Behavior as well as the relevant paper, i.e., Ulrich, Stefanec, Rekabi-bana et al.: Autonomous tracking of honeybee behaviors over long-term periods with cooperating robots. Science Robotics, 2024