Bee Swarm Activity Acoustic Classification for an IoT-Based Farm Service

Abstract

Beekeeping is one of the widespread and traditional fields in agriculture, where Internet of Things (IoT)-based solutions and machine learning approaches can ease and improve beehive management significantly. A particularly important activity is bee swarming. A beehive monitoring system can be applied for digital farming to alert the user via a service about the beginning of swarming, which requires a response. An IoT-based bee activity acoustic classification system is proposed in this paper. The audio data needed for acoustic training was collected from the Open Source Beehives Project. The input audio signal was converted into feature vectors, using the Mel-Frequency Cepstral Coefficients (with cepstral mean normalization) and Linear Predictive Coding. The influence of the acoustic background noise and denoising procedure was evaluated in an additional step. Different Hidden Markov Models' and Gaussian Mixture Models' topologies were developed for acoustic modeling, with the objective being to determine the most suitable one for the proposed IoT-based solution. The evaluation was carried out with a separate test set, in order to successfully classify sound between the normal and swarming conditions in a beehive. The evaluation results showed that good acoustic classification performance can be achieved with the proposed system.In the case of mobile networks, the 5G, and the already existing LTE-M1/NB-IoT, enable ubiquitous connectivity, which is important when the system gets distributed over more than one location. The increased number of connected devices, many of them originating from wireless-distributed sensor networks, were also responsible for the last major changes in network protocol standardization, where IPv6 was defined and deployed to communication networks, with the objective to cope with the new situation.If IoT and intelligent ambiance present one of the upfront sectors, agriculture is still a more traditional one. ICT technologies were introduced widely [6] to agriculture in the last two decades, but there is a large area of possibilities on how, additionally, to include real-time monitoring, machine learning and sensor networks into this sector. Sensor networks enable numerous solutions for smart farming [7], intending to increase the production rate and improve sustainability, but they are also, in parallel, in charge of taking care of animal condition and health. Beekeeping is an important part of agriculture, where the distributed locations are frequently present, and thus, indicate the need to ease the monitoring of animals in a 24/7 mode, which can benefit from advanced intelligent ambiance technologies. The popularity of modern urban beekeeping initiatives has gained further momentum to this research field, and presents an additional possibility as to how to include digital agriculture into a smart city and smart home solutions [8].Swarming has always been seen by beekeepers as an extremely important event, which requires an immediate response. Beekeepers were used to observing the beeh...

The prediction of swarming in honeybee colonies using vibrational spectra

“…Woods 18 ). In a recent paper by Zgank 21 , an Internet of Things based bee activity acoustic classification system is presented, with acoustic training data being collected from the Open Source Beehives Project. However, in contrast to this paper, which aims to predict whether a honeybee colony is preparing to swarm or not, their work focused only on identifying the best machine learning algorithm for acoustic classification between the two states, preparing and not preparing.…”

“…However, in contrast to this paper, which aims to predict whether a honeybee colony is preparing to swarm or not, their work focused only on identifying the best machine learning algorithm for acoustic classification between the two states, preparing and not preparing. Owing to the use of training data originating from only one hive, this categorisation would not be suitable for application across multiple hives 21 . Outside of the focus of swarming, acoustic measurements have also been explored to make assessments into the general health status of honeybee colonies.…”

Application of A Precision Apiculture System to Monitor Honey Daily Production

“…The honey production cycle takes place inside beehives placed in an open field in the presence of plants in bloom. This depends on many factors, some of which are environmental factors, such as temperature, relative humidity, and wind [2,4], and can benefit from advanced intelligent ambiance technologies [5].…”

“…A bee swarm monitoring system, operating with audio signals captured in a beehive, is described in [5]; the captured beehive audio data contain different bee activities, based on different frequencies and energies of the signal. These were studied to carry out a bee swarm activity acoustic classification useful to improve beehive management.…”

Design, Implementation and Simulation of a Fringing Field Capacitive Humidity Sensor

“…IoT applications have been reported to be used in healthcare, health monitoring and indoor tour guide systems [ 11 , 12 ]. Furthermore, the Internet of Things includes applications such as a bee activity acoustic classification system [ 13 ], a smartphone irrigation schedule tool supporting a farmer’s decision on the timing and amounts of irrigation [ 14 ] and a WSN monitoring the dosimetry of IRad operators [ 15 ]. IoT applications have evolved rapidly over the last few years and are now moving successfully to agriculture.…”