Imaging Stitching vs. Wider Field of View (FoV) Microscope Objective
Microscopy devices are designed for rather tight, niche applications, so what happens if your sample is bigger than your microscope allows? You can use mage stitching to remedy this issue or simply use a wider field of view. Each of these methods has its unique benefits, so neither answer is explicitly right or wrong.
We will examine both of them to help you understand which solution works better for your unique application. This includes looking at the latest image stitching technology, how optical lens elements impact the field of view, and what upcoming solutions might further improve the microscopic field of view.
What is Image Stitching?
Image stitching is simply the process of capturing multiple images and then stitching them together to form one wider image. Simplified versions of this technology exist in most modern smartphones to be used for taking panorama photos.
But if you’ve ever used this feature, you’ve seen how it can be prone to errors, and so it wouldn’t work with enough accuracy for a microscopy application. Instead, several dedicated open-source tools are developed to help grant this ability to researchers, including MIST.
The Microscopy Image Stitching Tool, or MIST, was introduced in July of 2017 by several researchers who published a piece on it in Scientific Reports.
They found that existing 2D stitching tools were not meeting the requirements needed for their projects, so they worked to create a time-lapsing mosaic from a grip of 2D overlapping images.
This vertical and horizontal overlap enabled consistent syncing between tiles through a three-step process. This process involved easing the transitions between the tiles, adjusting them to reduce errors in the overall image, and then composing the image into a finalized product.
They adopted a Fourier-based implementation to compute these transitions, as it is a simple and predictable solution that is enabled by the number of pixels captured from a high-definition objective lens.
The alternative would be a feature-based solution that requires a specific feature extraction step to detect common features in any two adjacent images. To help improve translations, they enabled MIST to estimate the mechanical factors of the lenses capturing the image to help understand spacing between them or the space a sample moved between photos.
Finally, to construct the mosaic, they built an algorithm to help optimize the approach, built on a minimum spanning tree framework to keep the image constrained.
The result is a stitching algorithm that utilizes both a CPU and GPU from a computer to calculate image positioning and properly stitch microscopy images together, which can be a challenge with other implementations.
Sparsely popular samples can lead to spatial gaps between frames, but MIST can navigate these situations to construct a finalized image correctly. This was done faster and with less space than other similar tools like TeraStitcher, iStitch, and multiple versions of FijilIS.
Currently, MIST is available within ImageJ at the National Institute of Standards and Technology’s website.
How Do I Calculate My Field of View (FoV) Number?
In microscopy, the field of view in a microscope is quantified as a field-of-view number or just a field number. Importantly, this is commonly understood as a concept that primarily concerns a microscope’s eyepiece, as objective lenses are focused and intentionally lacking a field of view. But, it is important for a researcher to understand the full scale of what they see through a microscope’s eyepiece and if its field of view is maximized.
We can understand this number by following MicroscopyU’s formula of multiplying our field size by our objective’s magnification, or if there is an auxiliary lens between the objective and the eyepiece, then this variable should be multiplied by the objective magnification as well.
However, it is important to consider that this field of view is limited by the objective lens and the size of the eyepiece, which can sometimes be limited. Older microscope models were limited to just 18 millimeters of usable field diameter, but more modern devices can accommodate 28 millimeters or even more in extreme cases.
Typically an eyepiece will have a field number of between 6 and 28 millimeters, depending on the magnification. Interestingly, lower magnification will often grant a larger field of view, so while a 10x eyepiece will offer between 16 and 18 millimeters in field of view, a 5x eyepiece will offer 20 millimeters. This is because the lesser optics can see more at the cost of magnification.
Limits and Variations of FoV
The piece within the eyepiece that can limit the field of view is known as the field diaphragm. Depending on your eyepiece’s design, it can either be constructed before the optical system entirely or between certain lenses.
These two variations are known as Huygens and Ramsden eyepieces, the former places the field lens first, and the latter places the field lens between the diaphragm and the eye lens.
In either case, the diaphragm is used to limit and control the flow of light, to prevent rays of light from reflecting off the side of the eyepiece and through the lens multiple times, making the edge of the image appear less clear.
This ability means that most microscopes include adjustable field of view numbers, and so these should be documented at the time of capturing an image or making an analysis.
Some modern eyepieces have worked to correct some of these lighting artifacts and can achieve field numbers of 26 millimeters or greater through what is referred to as wide-field designs.
These are still being developed and iterated on but present exciting solutions to view larger specimens simultaneously.
Will Microscopes Be Able To Reach Higher FoVs?
In November of 2020, a group of Chinese scientists published their findings on the use of metasurfaces in microscopy in SPIE’s Advanced Photonics. They found that with nanofabrication technology, they were able to create surfaces that could manipulate light while also retaining it, known as metalenses. The imaging performance can improve efficiency, a field of view, and even polarization, all with an ultrathin, ultralight, and flat architecture.
In their exercise, they went as far as to directly integrate the metalens into a CMOS image sensor, creating something they called a Metalens-Integrated Imaging Device or MIID. This device successfully broke the limitations of shadow imaging and its inability to capture the depth of field and was configured in an array to utilize multiple CMOS image sensors and conduct wide-field imaging.
This array also utilized a technique to accomplish the same objective as image stitching called polarization-multiplexed dual-phase, which was able to unify the image data at the time of capture. The result is high-resolution images with millimeter-scale from miniature device prototypes, which could be expanded in resolution, field of view, and depth of field.
Opportunities of Improved FoV
While this disruptive new technology is still in its infancy, it presents a lot of exciting opportunities to enable us to rethink microscopy and imaging technologies.
With these metasurfaces, it is seemingly possible to conduct several instances of microscopic imaging simultaneously, under extremely constrained conditions. Of course, there is still much distance to go here to understand the “how” behind this technology and how to implement it into current technology workflows for research purposes properly.
It states that microscopic imaging will only continue to increase, whether that be in resolution, the field of view, depth of field, or other metrics. With this new technology, we should look forward to new methods for conducting microscopy, new subjects and means of imaging, and other exciting opportunities that we have not even thought of yet.
The means for capturing a field of view in microscopy are complex, and there are several approaches to take depending on your application.
Image stitching uses algorithms to properly arrange several images into one concrete picture for scientific analysis. At the same time, widening the field of view through optical lenses and the field diaphragm is a possibility but can sometimes lessen magnification or quality in light received.
On the horizon are exciting new technologies that have the potential to change microscopy, and current solutions are certainly suitable for meeting the needs of researchers around the globe, even if there is room for improvement in many key aspects.
Scientific Reports - MIST: Accurate and Scalable Microscopy Image Stitching Tool with Stage Modeling and Error Minimization | Nature.org
MicroscopyU - Field of View | Microscopyu.com
SPIE Digital Library - Metalens-integrated compact imaging devices for wide-field microscopy | Spiedigitallibrary.org