Search

A continuing trend in precision apiculture is to use computer vision methods to quantify characteristics of bee traffic in managed colonies at the hive's entrance. Since traffic at the hive's entrance is a contributing factor to the hive's productivity and health, we assessed the potential of three open-source convolutional network models, YOLOv3, YOLOv4-tiny, and YOLOv7-tiny, to quantify omnidirectional traffic in videos from on-hive video loggers on regular, unmodified one- and two-super Langstroth hives and compared their accuracies, energy efficacies, and operational energy footprints. We trained and tested the models with a 70/30 split on a dataset of 23,173 flying bees manually labeled in 5819 images from 10 randomly selected videos and manually evaluated the trained models on 3600 images from 120 randomly selected videos from different apiaries, years, and queen races. We designed a new energy efficacy metric as a ratio of performance units per energy unit required to make a model operational in a continuous hive monitoring data pipeline. In terms of accuracy, YOLOv3 was first, YOLOv7-tiny—second, and YOLOv4-tiny—third. All models underestimated the true amount of traffic due to false negatives. YOLOv3 was the only model with no false positives, but had the lowest energy efficacy and highest operational energy footprint in a deployed hive monitoring data pipeline. YOLOv7-tiny had the highest energy efficacy and the lowest operational energy footprint in the same pipeline. Consequently, YOLOv7-tiny is a model worth considering for training on larger bee datasets if a primary objective is the discovery of non-invasive computer vision models of traffic quantification with higher energy efficacies and lower operational energy footprints. [ABSTRACT FROM AUTHOR]

Since bee traffic is a contributing factor to hive health and electromagnetic radiation has a growing presence in the urban milieu, we investigate ambient electromagnetic radiation as a predictor of bee traffic in the hive's vicinity in an urban environment. To that end, we built two multi-sensor stations and deployed them for four and a half months at a private apiary in Logan, UT, USA. to record ambient weather and electromagnetic radiation. We placed two non-invasive video loggers on two hives at the apiary to extract omnidirectional bee motion counts from videos. The time-aligned datasets were used to evaluate 200 linear and 3,703,200 non-linear (random forest and support vector machine) regressors to predict bee motion counts from time, weather, and electromagnetic radiation. In all regressors, electromagnetic radiation was as good a predictor of traffic as weather. Both weather and electromagnetic radiation were better predictors than time. On the 13,412 time-aligned weather, electromagnetic radiation, and bee traffic records, random forest regressors had higher maximum R 2 scores and resulted in more energy efficient parameterized grid searches. Both types of regressors were numerically stable. [ABSTRACT FROM AUTHOR]

The relationship between beehive weight and traffic is a fundamental open research problem for electronic beehive monitoring and digital apiculture, because weight and traffic affect many aspects of honeybee (Apis mellifera) colony dynamics. An investigation of this relationship was conducted with a nondisruptive two-sensor (scale and camera) system on the weight and video data collected on six Apis mellifera colonies in Langstroth hives at the USDA-ARS Carl Hayden Bee Research Center in Tucson, Arizona, USA, from 15 May to 15 August 2021. Three hives had positive and two hives had negative correlations between weight and traffic. In one hive, weight and traffic were uncorrelated. The strength of the correlation between weight and traffic was stronger for longer time intervals. The traffic spread and mean, when taken separately, did not affect the correlation between weight and traffic more significantly than the exact traffic counts from videos. Lateral traffic did not have a significant impact on weight. [ABSTRACT FROM AUTHOR]

Omnidirectional honeybee traffic is the number of bees moving in arbitrary directions in close proximity to the landing pad of a beehive over a period of time. Automated video analysis of such traffic is critical for continuous colony health assessment. In our previous research, we proposed a two-tier algorithm to measure omnidirectional bee traffic in videos. Our algorithm combines motion detection with image classification: in tier 1, motion detection functions as class-agnostic object location to generate regions with possible objects; in tier 2, each region from tier 1 is classified by a class-specific classifier. In this article, we present an empirical and theoretical comparison of random reinforced forests and shallow convolutional networks as tier 2 classifiers. A random reinforced forest is a random forest trained on a dataset with reinforcement learning. We present several methods of training random reinforced forests and compare their performance with shallow convolutional networks on seven image datasets. We develop a theoretical framework to assess the complexity of image classification by a image classifier. We formulate and prove three theorems on finding optimal random reinforced forests. Our conclusion is that, despite their limitations, random reinforced forests are a reasonable alternative to convolutional networks when memory footprints and classification and energy efficiencies are important factors. We outline several ways in which the performance of random reinforced forests may be improved. [ABSTRACT FROM AUTHOR]

Accurate measurement of honeybee (Apis mellifera) traffic in the vicinity of the hive is critical in systems that continuously monitor honeybee colonies to detect deviations from the norm. BeePIV, the algorithm we describe and evaluate in this article, is a new significant result in our longitudinal investigation of honeybee flight and traffic in electronic beehive monitoring. BeePIV converts frames from bee traffic videos to particle motion frames with uniform background, applies particle image velocimetry to these motion frames to compute particle displacement vector fields, classifies individual displacement vectors as incoming, outgoing, and lateral, and uses the respective vector counts to measure incoming, outgoing, and lateral bee traffic. We evaluate BeePIV on twelve 30-s color videos with a total frame count of 8928 frames for which we obtained the ground truth by manually counting every full bee motion in each frame. The bee motion counts obtained from these videos with BeePIV come closer to the human bee motion counts than the bee motion counts obtained with our previous video-based bee counting methods. We use BeePIV to compute incoming and outgoing bee traffic curves for two different hives over a period of seven months and observe that these curves closely follow each other. Our observations indicate that bee traffic curves obtained by BeePIV may be used to predict colony failures. Our experiments suggest that BeePIV can be used in situ on the raspberry pi platform to process bee traffic videos. Dataset: The supplementary materials for our article include three video sets for the reader to appreciate how BeePIV processes bee traffic videos. Each set consists of three videos. The first video is an original raw video captured by our deployed BeePi electronic beehive monitoring system; the second video is the corresponding video that consists of the white background particle motion frames extracted from the first video; the third video is the video of displacement vectors extracted by PIV from each pair of consecutive white background frames from the second video. All computation is executed on a raspberry pi 3 model B v1.2 with four cores. [ABSTRACT FROM AUTHOR]

The well-being of a honeybee (Apis mellifera) colony depends on forager traffic. Consistent discrepancies in forager traffic indicate that the hive may not be healthy and require human intervention. Honeybee traffic in the vicinity of a hive can be divided into three types: incoming, outgoing, and lateral. These types constitute directional traffic, and are juxtaposed with omnidirectional traffic where bee motions are considered regardless of direction. Accurate measurement of directional honeybee traffic is fundamental to electronic beehive monitoring systems that continuously monitor honeybee colonies to detect deviations from the norm. An algorithm based on digital particle image velocimetry is proposed to measure directional traffic. The algorithm uses digital particle image velocimetry to compute motion vectors, analytically classifies them as incoming, outgoing, or lateral, and returns the classified vector counts as measurements of directional traffic levels. Dynamic time warping is used to compare the algorithm's omnidirectional traffic curves to the curves produced by a previously proposed bee motion counting algorithm based on motion detection and deep learning and to the curves obtained from a human observer's counts on four honeybee traffic videos (2976 video frames). The currently proposed algorithm not only approximates the human ground truth on par with the previously proposed algorithm in terms of omnidirectional bee motion counts but also provides estimates of directional bee traffic and does not require extensive training. An analysis of correlation vectors of consecutive image pairs with single bee motions indicates that correlation maps follow Gaussian distribution and the three-point Gaussian sub-pixel accuracy method appears feasible. Experimental evidence indicates it is reasonable to treat whole bees as tracers, because whole bee bodies and not parts thereof cause maximum motion. To ensure the replicability of the reported findings, these videos and frame-by-frame bee motion counts have been made public. The proposed algorithm is also used to investigate the incoming and outgoing traffic curves in a healthy hive on the same day and on different days on a dataset of 292 videos (216,956 video frames). [ABSTRACT FROM AUTHOR]

Omnidirectional bee traffic is the number of bees moving in arbitrary directions in close proximity to the landing pad of a given hive over a given period of time. Video bee traffic analysis has the potential to automate the assessment of omnidirectional bee traffic levels, which, in turn, may lead to a complete or partial automation of honeybee colony health assessment. In this investigation, we proposed, implemented, and partially evaluated a two-tier method for counting bee motions to estimate levels of omnidirectional bee traffic in bee traffic videos. Our method couples motion detection with image classification so that motion detection acts as a class-agnostic object location method that generates a set of regions with possible objects and each such region is classified by a class-specific classifier such as a convolutional neural network or a support vector machine or an ensemble of classifiers such as a random forest. The method has been, and is being iteratively field tested in BeePi monitors, multi-sensor electronic beehive monitoring systems, installed on live Langstroth beehives in real apiaries. Deployment of a BeePi monitor on top of a beehive does not require any structural modification of the beehive's woodenware, and is not disruptive to natural beehive cycles. To ensure the replicability of the reported findings and to provide a performance benchmark for interested research communities and citizen scientists, we have made public our curated and labeled image datasets of 167,261 honeybee images and our omnidirectional bee traffic videos used in this investigation. Dataset: The three image datasets (BEE1, BEE2_1S and BEE2_2S) used in this investigation are publicly available at https://usu.box.com/s/0a5yurmn7ija15cp236awa1re65mbcnr , https://usu.box.com/s/p7y8v95ot9u9jvjbbayci61no3lzbgx3 and https://usu.box.com/s/3ccizd5b1qzcqcs4t0ivawmrxbgva7ym , respectively. The omnidirectional bee traffic videos and the corresponding human bee motion counts are included in the Supplementary Materials. [ABSTRACT FROM AUTHOR]