An official website of the United States government
We present a novel system for the automatic video monitoring of honey bee foraging activity at the hive entrance. This monitoring system is built upon convolutional neural networks that perform multiple animal pose estimation without the need for marking. This precise detection of honey bee body parts is a key element of the system to provide detection of entrance and exit events at the entrance of the hive including accurate pollen detection. A detailed evaluation of the quality of the detection and a study of the effect of the parameters are presented. The complete system also integrates identification of barcode marked bees, which enables the monitoring at both aggregate and individual levels. The results obtained on multiple days of video recordings show the applicability of the approach for large-scale deployment. This is an important step forward for the understanding of complex behaviors exhibited by honey bees and the automatic assessment of colony health.
more »
« less
Analysis of Honeybee Drone Activity during the Mating Season in Northwestern Argentina
Males in Hymenopteran societies are understudied in many aspects and it is assumed that they only have a reproductive function. We studied the time budget of male honey bees, drones, using multiple methods. Changes in the activities of animals provide important information on biological clocks and their health. Yet, in nature, these changes are subtle and often unobservable without the development and use of modern technology. During the spring and summer mating season, drones emerge from the hive, perform orientation flights, and search for drone congregation areas for mating. This search may lead drones to return to their colony, drift to other colonies (vectoring diseases and parasites), or simply get lost to predation. In a low percentage of cases, the search is successful, and drones mate and die. Our objective was to describe the activity of Apis mellifera drones during the mating season in Northwestern Argentina using three methods: direct observation, video recording, and radio frequency identification (RFID). The use of RFID tagging allows the tracking of a bee for 24 h but does not reveal the detailed activity of drones. We quantified the average number of drones’ departure and arrival flights and the time outside the hive. All three methods confirmed that drones were mostly active in the afternoon. We found no differences in results between those obtained by direct observation and by video recording. RFID technology enabled us to discover previously unknown drone behavior such as activity at dawn and during the morning. We also discovered that drones may stay inside the hive for many days, even after initiation of search flights (up to four days). Likewise, we observed drones to leave the hive for several days to return later (up to three days). The three methods were complementary and should be considered for the study of bee drone activity, which may be associated with the diverse factors influencing hive health.
more »
« less
- PAR ID:
- 10310371
- Date Published:
- Journal Name:
- Insects
- Volume:
- 12
- Issue:
- 6
- ISSN:
- 2075-4450
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Cheapcommercial off-the-shelf (COTS)First-Person View (FPV)drones have become widely available for consumers in recent years. Unfortunately, they also provide low-cost attack opportunities to malicious users. Thus, effective methods to detect the presence of unknown and non-cooperating drones within a restricted area are highly demanded. Approaches based on detection of drones based on emitted video stream have been proposed, but were not yet shown to work against other similar benign traffic, such as that generated by wireless security cameras. Most importantly, these approaches were not studied in the context of detecting new unprofiled drone types. In this work, we propose a novel drone detection framework, which leverages specific patterns in video traffic transmitted by drones. The patterns consist of repetitive synchronization packets (we call pivots), which we use as features for a machine learning classifier. We show that our framework can achieve up to 99% in detection accuracy over an encrypted WiFi channel using only 170 packets originated from the drone within 820ms time period. Our framework is able to identify drone transmissions even among very similar WiFi transmissions (such as video streams originated from security cameras) as well as in noisy scenarios with background traffic. Furthermore, the design of our pivot features enables the classifier to detect unprofiled drones in which the classifier has never trained on and is refined using a novel feature selection strategy that selects the features that have the discriminative power of detecting new unprofiled drones.more » « less
-
Daniel, Sloan (Ed.)Abstract Recent declines in the health of the honey bee have startled researchers and lay people alike as honey bees are agriculture’s most important pollinator. Honey bees are important pollinators of many major crops and add billions of dollars annually to the US economy through their services. One factor that may influence colony health is the microbial community. Indeed, the honey bee worker digestive tract harbors a characteristic community of bee-specific microbes, and the composition of this community is known to impact honey bee health. However, the honey bee is a superorganism, a colony of eusocial insects with overlapping generations where nestmates cooperate, building a hive, gathering and storing food, and raising brood. In contrast to what is known regarding the honey bee worker gut microbiome, less is known of the microbes associated with developing brood, with food stores, and with the rest of the built hive environment. More recently, the microbe Bombella apis was identified as associated with nectar, with developing larvae, and with honey bee queens. This bacterium is related to flower-associated microbes such as Saccharibacter floricola and other species in the genus Saccharibacter, and initial phylogenetic analyses placed it as sister to these environmental bacteria. Here, we used comparative genomics of multiple honey bee-associated strains and the nectar-associated Saccharibacter to identify genomic changes that may be associated with the ecological transition to honey bee association. We identified several genomic differences in the honey bee-associated strains, including a complete CRISPR/Cas system. Many of the changes we note here are predicted to confer upon Bombella the ability to survive in royal jelly and defend themselves against mobile elements, including phages. Our results are a first step toward identifying potential function of this microbe in the honey bee superorganism.more » « less
-
UAVs (unmanned aerial vehicles) or drones are promising instruments for video-based surveillance. Various applications of aerial surveillance use object detection programs to detect target objects. In such applications, three parameters influence a drone deployment strategy: the area covered by the drone, the latency of target (object) detection, and the quality of the detection output by the object detector. Previous works have focused on improving Pareto optimality along the area-latency frontier or the area-quality frontier, but not on the combined area-latency-quality frontier, because of which these solutions are sub-optimal for drone-based surveillance. We explore a three way tradeoff between area, latency, and quality in the context of autonomous aerial surveillance of targets in an area using drones with cameras and an object detection program. We propose Vega, a drone deployment framework that captures these tradeoffs to deploy drones efficiently. We make three contributions with Vega. First, we characterize the ability of the state-of-the-art mobile object detector, EfficientDet [CPVR '20], to detect objects from varying drone altitudes using confidence and IoU curves vs. drone altitude. Second, based on these characteristics of the detector, we propose a set of two algorithmic primitives for drone-based maneuvers, namely DroneZoom and DroneCycle. Using these two primitives, we obtain a more optimal Pareto frontier between our three target parameters - coverage area, detection latency, and detection quality for a single drone system. Third, we scale out our findings to a swarm deployment using higher-order Voronoi tessellations, where we control the swarm's spatial density using the Voronoi order to further lower the detection latency while maintaining detection quality.more » « less
-
he pervasive operation of customer drones, or small-scale unmanned aerial vehicles (UAVs), has raised serious concerns about their privacy threats to the public. In recent years, privacy invasion events caused by customer drones have been frequently reported. Given such a fact, timely detection of invading drones has become an emerging task. Existing solutions using active radar, video or acoustic sensors are usually too costly (especially for individuals) or exhibit various constraints (e.g., requiring visual line of sight). Recent research on drone detection with passive RF signals provides an opportunity for low-cost deployment of drone detectors on commodity wireless devices. However, the state of the arts in this direction rely on line-of-sight (LOS) RF signals, which makes them only work under very constrained conditions. The support of more common scenarios, i.e., non-line-of-sight (NLOS), is still missing for low-cost solutions. In this paper, we propose a novel detection system for privacy invasion caused by customer drone. Our system is featured with accurate NLOS detection with low-cost hardware (under $50). By exploring and validating the relationship between drone motions and RF signal under the NLOS condition, we find that RF signatures of drones are somewhat “amplified” by multipaths in NLOS. Based on this observation, we design a two-step solution which first classifies received RSS measurements into LOS and NLOS categories; deep learning is then used to extract the signatures and ultimately detect the drones. Our experimental results show that LOS and NLOS signals can be identified at accuracy rates of 98.4% and 96% respectively. Our drone detection rate for NLOS condition is above 97% with a system implemented using Raspberry PI 3 B+.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/insects12060566
