Automatic methods for long-term tracking and the detection and decoding of communication dances in honeybees

Wario, Fernando; Wild, Benjamin; Couvillon, Margaret J.; Rojas, Raúl; Landgraf, Tim

doi:10.3389/fevo.2015.00103

Cited by 57 publications

(58 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A decoder DCNN trained on this data outperforms decoders trained on various baselines. Compared to the previously used non-neural algorithm (Wario et al, 2015) RenderGAN improved the decoding accuracy significantly. Consequently the downstream tracking process that links detections through time improved in several respects.…”

Section: Discussionmentioning

confidence: 92%

See 1 more Smart Citation

RenderGAN: Generating Realistic Labeled Data

2018

Self Cite

View full text Add to dashboard Cite

Deep Convolutional Neuronal Networks (DCNNs) are showing remarkable performance on many computer vision tasks. Due to their large parameter space, they require many labeled samples when trained in a supervised setting. The costs of annotating data manually can render the use of DCNNs infeasible. We present a novel framework called RenderGAN that can generate large amounts of realistic, labeled images by combining a 3D model and the Generative Adversarial Network framework. In our approach, image augmentations (e.g., lighting, background, and detail) are learned from unlabeled data such that the generated images are strikingly realistic while preserving the labels known from the 3D model. We apply the RenderGAN framework to generate images of barcode-like markers that are attached to honeybees. Training a DCNN on data generated by the RenderGAN yields considerably better performance than training it on various baselines.

show abstract

Section: Discussionmentioning

confidence: 92%

“…The previous BeesBook vision system (Wario et al, 2015) was tailored to specifically track all animals of small honeybee colonies over their entire lifetime. It uses a round and curved marker and searches for ellipse-shaped edge formations.…”

Section: Related Workmentioning

confidence: 99%

RenderGAN: Generating Realistic Labeled Data

2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…Providing a broad set of insightful tools, only highly integrated research on novel systems leads to a meaningful design method for hybrid societies. Prime examples of successful integration of methods are the integration of robots and fish (Marras and Porfiri, 2012) and the automatic analysis of social networks in honeybees (Wario et al, 2015). Again, establishing such a deep understanding of the other field requires time.…”

Section: Secondary Challenge: Integration Of Methodsmentioning

confidence: 99%

Hybrid Societies: Challenges and Perspectives in the Design of Collective Behavior in Self-organizing Systems

et al. 2016

View full text Add to dashboard Cite

Hybrid societies are self-organizing, collective systems, which are composed of different components, for example, natural and artificial parts (bio-hybrid) or human beings interacting with and through technical systems (socio-technical). Many different disciplines investigate methods and systems closely related to the design of hybrid societies. A stronger collaboration between these disciplines could allow for re-use of methods and create significant synergies. We identify three main areas of challenges in the design of self-organizing hybrid societies. First, we identify the formalization challenge. There is an urgent need for a generic model that allows a description and comparison of collective hybrid societies. Second, we identify the system design challenge. Starting from the formal specification of the system, we need to develop an integrated design process. Third, we identify the challenge of interdisciplinarity. Current research on self-organizing hybrid societies stretches over many different fields and hence requires the re-use and synthesis of methods at intersections between disciplines. We then conclude by presenting our perspective for future approaches with high potential in this area.

show abstract

“…A natural honey bee colony contains a high density of visually similar members, rapidly moving and occluding on the uneven and changing honeycomb surface, and whose numbers change in time. These factors present substantial difficulties for automated image analysis for which a common solution is to apply physical tags to some 8 or most 9,10 of the colony members. Barcoded tags allow for the distinct marking of a sufficiently large number of individuals to track a naturally-sized colony and have been exploited to unravel important aspects of bee communication 8,11 and information spread 10 .…”

Section: Introductionmentioning

confidence: 99%

“…However, the burden of manually tagging hundreds or thousands of small insects, without harm or inhibition to their motion, carries severe limitations. For example, marking newly hatched bees requires either opening the hive or introducing marked newborns without letting any hatch inside, both of which disrupt the colony 9,11 . Moreover, the recognition of markers fails when the tag is partially occluded or blurred.…”

Section: Introductionmentioning

confidence: 99%

Markerless tracking of an entire insect colony

Bożek

Hebert

Portugal

et al. 2020

Preprint

View full text Add to dashboard Cite

We present a comprehensive, computational method for tracking an entire colony of the honey bee Apis mellifera using high-resolution video on a natural honeycomb background. We adapt a convolutional neural network (CNN) segmentation architecture to automatically identify bee and brood cell positions, body orientations and within-cell states. We achieve high accuracy (~10% body width error in position, ~10° error in orientation, and true positive rate > 90%) and demonstrate months-long monitoring of sociometric colony fluctuations. We combine extracted positions with rich visual features of organismcentered images to track individuals over time and through challenging occluding events, recovering ~79% of bee trajectories from five observation hives over a span of 5 minutes. The resulting trajectories reveal important behaviors, including fast motion, comb-cell activity, and waggle dances. Our results provide new opportunities for the quantitative study of collective bee behavior and for advancing tracking techniques of crowded systems.

show abstract

Automatic methods for long-term tracking and the detection and decoding of communication dances in honeybees

Abstract: Automatic methods for long-term tracking and the detection and decoding of communication dances in honeybees.

Cited by 57 publications

References 51 publications

RenderGAN: Generating Realistic Labeled Data

RenderGAN: Generating Realistic Labeled Data

Hybrid Societies: Challenges and Perspectives in the Design of Collective Behavior in Self-organizing Systems

Markerless tracking of an entire insect colony

Contact Info

Product

Resources

About