%0 Conference Paper %1 10180253 %A Janetzky, Pascal %A Schaller, Melanie %A Krause, Anna %A Hotho, Andreas %B 2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP) %D 2023 %K author:KRAUSE from:jascal_panetzky machine-learning myown we4bee we4bee_audio %P 1-5 %R 10.1109/IWSSIP58668.2023.10180253 %T Swarming Detection in Smart Beehives Using Auto Encoders for Audio Data %X —Swarming is the natural mechanism by which bee colonies reproduce, but for beekeepers it is a challenge. Precision beekeeping can aid their work through early notifications about impending swarms. In this work, we focus on identifying swarms and their early indicators in audio data captured from a smart beehive. The challenge with such domain-specific data is the low availability of labelled samples, the strong label imbalance, and the recording of undesired sources. We approach this challenge through a two-step setup: First, we use an auto encoder network to detect sounds from mechanical sources and then use it to clean data. Secondly, on the cleaned data we then employ a second network to identify event-related bee sounds. Using spectrogram features, our networks are able to reach a balanced accuracy score of more than 99 % in the detection of special bee events. The findings of this initial study can serve as the starting point for further research on handling imbalanced data collections from smart, remote sensor environments that also contain undesired signals.

@inproceedings{10180253, abstract = {—Swarming is the natural mechanism by which bee colonies reproduce, but for beekeepers it is a challenge. Precision beekeeping can aid their work through early notifications about impending swarms. In this work, we focus on identifying swarms and their early indicators in audio data captured from a smart beehive. The challenge with such domain-specific data is the low availability of labelled samples, the strong label imbalance, and the recording of undesired sources. We approach this challenge through a two-step setup: First, we use an auto encoder network to detect sounds from mechanical sources and then use it to clean data. Secondly, on the cleaned data we then employ a second network to identify event-related bee sounds. Using spectrogram features, our networks are able to reach a balanced accuracy score of more than 99 % in the detection of special bee events. The findings of this initial study can serve as the starting point for further research on handling imbalanced data collections from smart, remote sensor environments that also contain undesired signals.}, added-at = {2023-08-11T03:18:08.000+0200}, author = {Janetzky, Pascal and Schaller, Melanie and Krause, Anna and Hotho, Andreas}, biburl = {https://www.bibsonomy.org/bibtex/258cc30c189bcf7d38fcbbf50c031a85e/dmir}, booktitle = {2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP)}, doi = {10.1109/IWSSIP58668.2023.10180253}, interhash = {6ffc8218c5defb7a76d5cdfed678ce95}, intrahash = {58cc30c189bcf7d38fcbbf50c031a85e}, keywords = {author:KRAUSE from:jascal_panetzky machine-learning myown we4bee we4bee_audio}, pages = {1-5}, timestamp = {2024-01-18T10:31:52.000+0100}, title = {Swarming Detection in Smart Beehives Using Auto Encoders for Audio Data}, year = 2023 }