Levitating Bugs With Sound Could Transform Scientific Photography

Scientists used precisely controlled soundwaves to suspend insect specimens, capturing detailed photographs from multiple angles without damaging the specimens.
Macro photography of small specimens is a routine part of scientific research. However, the typical approach of “posing” subjects using pins can damage sensitive specimens and is not an option at all for the smallest specimens.
Researchers from the Institute for Automation and Applied Infomatics, Karlsruhe Institute of Technology; the Center for Integrative Biodiversity Discovery, Leibniz Institute for Evolution and Biodiversity Science; and the Institute for Biology, Humboldt University Berlin, utilized sound waves to overcome these limitations.
As Hackaday notes, acoustic levitation is not itself novel. This method for suspending lightweight matter in air using high-intensity sound waves has been around since the early 20th century and continues to interest scientists. However, the practical application of acoustic levitation for something like scientific photography is new and fascinating.
The German researchers showed how acoustic levitation can work for insect specimens and demonstrated an automated, controllable way to maneuver specimens to capture specific images at pre-defined angles. Further, they can perform automated focus stacking on specimens, as they are kept steady enough, and capture highly detailed photos with an extended depth of field. It is also possible to turn this expansive dataset into highly detailed 3D models of specimens without hands-on repositioning, manual image capture, or any risk of damaging the specimen.

Insects are an especially crucial subject of study in biodiversity research. As the researchers explain, insects contribute “more than 75% of the species-level biodiversity” observed and are vital to ecosystem functions. However, “a major obstacle to studying insects” is that they are not easy to identify or study. Machine learning is making identification much less challenging, but training these AI models requires a lot of data. Collecting this detailed data about individual specimens requires taking many photos, which, as described above, is hard. Cost-effective, accessible, and repeatable imaging systems for collecting the necessary photos are critical.
While the science of acoustic levitation is fascinating, and detailed in the complete research paper, it is enough at this point to say that it is possible to control the position of the specimens with enough consistency and precision to capture a series of 40 photographs per specimen location. While the team admits that it did not need all 40 shots to achieve the desired depth of field after focus stacking, it is better to have too many images than not enough. They repeated this process for 72 different perspectives by moving the specimen while maintaining a stationary camera. Focus stacking was performed using Helicon Focus software, a popular choice for many macro photographers.
Speaking of gear, the researchers used an Olympus OM-D E-M1 Mark III camera with an M.Zuiko Digital ED 90mm f/3.5 Macro lens plus a 2x teleconverter. Like Helicon Focus, this camera and lens setup is popular among macro photography enthusiasts.

“Unlike other photogrammetric imaging systems used in biodiversity research, the object is not manipulated using an insect pin or other physical attachments,” the researchers write. “This ensures the specimen is not physically damaged, and since the imaging is sufficiently quick in comparison to previous systems the specimens do not dry to such an extent that they float on ethanol after returning them to a vial. This means that further work on the NDA (genome sequencing) is facilitated. Furthermore, all parts of the specimen can be viewed as no pin is blocking the view of the specimen. It also ensures insects too small to be pinned can be imaged, and the time and labor-intensive pinning process is avoided.”
While the team admits limitations, including that acoustic field rotation is limited to a single axis for now and that a perfect 3D model has not yet been achieved, the approach has significant potential.
“The proposed automated imaging system offers a promising solution for close-range photogrammetry of small, light objects. It is a non-destructive approach to multiperspective imaging and offers the basis for 3D reconstruction,” the researchers conclude.
Image credits: ‘Automated Photogrammetric Close-Range Imaging System for Small Invertebrates Using Acoustic Levitation‘ by Nathalie Klug, Markus Kramer, Fahri Mazrek, Lorenz Wührl, Hossein Shirali, Rudolf Meier, and Christian Pylaiuk
Source link