New method by University of Tartu scientists turns foggy holograms into a clear image
Researchers at the University of Tartu's Institute of Physics have developed a new technique to improve the quality of three-dimensional images. The method allows computational imaging systems to increase the depth of field of holograms — the portion that appears in focus — by up to five times after the fact.
One of the main drawbacks of traditional microscopes and 3D imaging systems is that once an image or hologram has been captured, its focus characteristics can no longer be adjusted. To overcome this limitation, University of Tartu Institute of Physics doctoral student Shivasubramanian Gopinath and his colleagues have developed a new method.
Their solution captures not just a single image, but a set of holograms at different focal depths at the moment of imaging. These can then be computationally merged into a synthetic hologram. Figuratively speaking, the computer stitches together multiple images like a patchwork quilt. The resulting composite image is sharply focused across a much broader range than would be physically possible with a single capture. This kind of hologram offers significantly greater depth of field and is easier to process afterward.
The new method represents a major advancement over a previous smart capture technique for holograms. That earlier approach enabled the recording of an object's three-dimensional information in normal lighting conditions and allowed a computer to later reconstruct the spatial image — a process known as Fresnel incoherent correlation holography (FINCH).
The research team calls the new method "post-engineering of axial resolution in FINCH," or PEAR-FINCH. By applying it, 3D holographic microscopy becomes more reliable under challenging conditions and allows for easier study of complex biological structures.
What makes the new method unique?
According to its developers, the new method is unique for several reasons. First, it allows the depth of field to be modified after the hologram has been captured. Second, its two-step computational reconstruction process keeps the image sharp and prevents it from being lost in noise.
The research team also achieved a fivefold increase in depth of field compared to the standard FINCH method. This means scientists can view much thicker tissue samples or larger parts of a cell in sharp focus all at once, without constantly refocusing the microscope up and down. Finally, the method performs well even under diffuse lighting, which is typical when imaging biological samples.
"This kind of post-capture flexibility has never been seen before. So we can say our achievement represents a new paradigm in holographic imaging and consistently outperforms both conventional direct imaging systems and standard FINCH," said Shivasubramanian Gopinath.
Sharpness requires time
As part of their work, the research team conducted an experiment comparing the new method with both conventional direct imaging and existing holographic techniques such as FINCH. The results showed that when objects were positioned, for example, 12 millimeters apart, they appeared blurry using traditional methods. In contrast, PEAR-FINCH was able to render both objects sharply in focus at the same time.
While the new method offers unprecedented flexibility, it comes at the cost of greater time and data requirements. To produce a single image, the system must record and process significantly more information, capturing roughly three times as many frames as standard FINCH. As a result, the technique is not yet suitable for imaging very fast-moving objects.
The research findings were published in the journal Journal of Physics: Photonics in an article titled "Axial resolution post-processing engineering in Fresnel incoherent correlation holography."
--
Editor: Marcus Turovski, Airika Harrik
