The Electronic Bee Spy: Eavesdropping on Honeybee Communication via Electrostatic Field Recordings.

As a canary in a coalmine warns of dwindling breathable air, the honeybee can indicate the health of an ecosystem. Honeybees are the most important pollinators of fruit-bearing flowers, and share similar ecological niches with many other pollinators; therefore, the health of a honeybee colony can reflect the conditions of a whole ecosystem. The health of a colony may be mirrored in social signals that bees exchange during their sophisticated body movements such as the waggle dance. To observe these changes, we developed an automatic system that records and quantifies social signals under normal beekeeping conditions. Here, we describe the system and report representative cases of normal social behavior in honeybees. Our approach utilizes the fact that honeybee bodies are electrically charged by friction during flight and inside the colony, and thus they emanate characteristic electrostatic fields when they move their bodies. These signals, together with physical measurements inside and outside the colony (temperature, humidity, weight of the hive, and activity at the hive entrance) will allow quantification of normal and detrimental conditions of the whole colony. The information provided instructs how to setup the recording device, how to install it in a normal bee colony, and how to interpret its data.

Estimating and abstracting the 3D structure of feline bones using neural networks on X-ray (2D) images.

Computing 3D bone models using traditional Computed Tomography (CT) requires a high-radiation dose, cost and time. We present a fully automated, domain-agnostic method for estimating the 3D structure of a bone from a pair of 2D X-ray images. Our triplet loss-trained neural network extracts a 128-dimensional embedding of the 2D X-ray images. A classifier then finds the most closely matching 3D bone shape from a predefined set of shapes. Our predictions have an average root mean square (RMS) distance of 1.08 mm between the predicted and true shapes, making our approach more accurate than the average achieved by eight other examined 3D bone reconstruction approaches. Each embedding extracted from a 2D bone image is optimized to uniquely identify the 3D bone CT from which the 2D image originated and can serve as a kind of fingerprint of each bone; possible applications include faster, image content-based bone database searches for forensic purposes.