In today’s data-driven world, the ability to see beyond what human eyes can detect is changing how we understand everything from crop health to natural gas detection.
Hyperspectral imaging is gaining popularity thanks to its ability to capture and analyse a huge amount of spectral information. This can enable the non-invasive and non-destructive identification and quantification of materials and objects across various industries and research applications in industries ranging from agriculture to semiconductor manufacturing.
According to the latest report from ResearchAndMarkets.com, the hyperspectral imaging market was valued at $301.4 million last year, and is projected to reach $472.9 million by 2029, a compound annual growth rate of 9.40%.
This, says the Hyperspectral Imaging Market report, is driven by technological developments, such as the integration of hyperspectral imaging with machine learning (ML) techniques, to improve the accuracy and speed of freshness detection in perishable food items, for example. The report highlights that advances such as this could ensure the fast and non-destructive evaluation of food quality, so the food industry can reduce waste and maintain food safety.
The intelligence firm also highlighted in its findings how hyperspectral imaging technologies are being miniaturised and integrated into small satellite platforms, which has led to them being more prevalently used in environmental and agricultural applications.
What is hyperspectral imaging?
Conventional cameras capture energy in broad bands from the UV to Infrared. Hyperspectral imaging systems can collect data across dozens or even hundreds of narrow, adjacent wavelength bands. This provides a detailed spectral "fingerprint" that can reveal properties invisible to the naked eye.
"There's a lot of work that we do in the arena of hyperspectral imaging, supporting agriculture and machine vision, as well as the people that are doing quality control in those regions," explains Stephen Washkevich, Senior Application Engineer at Omega Optical. "One of the things that's most important is making sure that people are getting exactly the light that they need."
The technology relies on sophisticated optical filters that can isolate specific wavelengths of light. In what Washkevich describes as "multispectral imaging," separate detectors equipped with discrete filters capture different wavelengths simultaneously. But hyperspectral imaging employs an advanced solution: linear variable filters. Says Washkevich. "The way those work is over the surface of the filter, the centre wavelength, or the wavelength that it discriminates, changes. So now you have one filter that can serve the purpose of effectively 110 filters."
Agricultural revolution
As alluded to in the report, one of the most established applications for hyperspectral imaging is in agriculture, where aerial systems mounted on drones or planes can assess crop health across vast areas.
"In the agricultural realm, often, teams are looking for a number of specific indicators of the health of crops," Washkevich notes. "What we're doing is coming up with a series of very specific bandpass filters that cover wide angles of incidence, for drones or plane-based camera systems."
These systems can detect problems long before they become visible to farmers. "It gives them the opportunity to see very early on whether or not a specific crop is affected by some sort of blight, or if they are under- or over-watered," he explains.
Environmental monitoring
Beyond agriculture, hyperspectral imaging plays a key role in environmental monitoring, particularly in detecting gases and pollutants. "There are a lot of people that are looking for methane detection or natural gas deposits, and so those have certain absorption lines and certain spectral fingerprints," says Washkevich. By identifying these unique spectral signatures, hyperspectral systems can pinpoint leaks or map deposits with precision.
Space-based observation
Some of the more ambitious applications involve space-based platforms that collect huge amounts of spectral data about the planet. "Sometimes we receive requests for people that are working on space-based observation platforms," Washkevich says. "They'll want a linear variable filter, and they just want to collect as much information as they can each second that they're observing the earth."
This approach creates versatile datasets that can serve multiple purposes. "If you can collect as much information as possible, it becomes multi-purpose," adds Washkevich. “But that’s one of the big challenges, trying to service the people who want to collect all the information, rather than a specific set.”
Engineering challenges
Creating the optical components for hyperspectral imaging systems presents its own unique manufacturing challenges. Unlike conventional optical filters where uniformity is the goal, linear variable filters require precisely controlled variation.
"When we make a filter, we want the entire filter to do one thing very well and uniformly." explains Washkevich. "An LVF is almost the exact opposite of that. You want it to vary very specifically over an area."
The company has developed unique techniques to achieve this controlled variation. "The biggest part has to do with changing the way that we shape our deposition." he says. "We go through a complex process of modeling and shaping apertures in front of the coating cloud. And so what that does is give it a -linear shadow effect that affects the thickness of each one of those layers."
Once manufactured, measuring these specialized filters can present an additional challenge. "When you're measuring a standard filter, the whole filter has the same properties. The linear variable is changing continuously," Washkevich points out. "So, how do you go about identifying what it is and where it's doing that particular function?"
A collaborative approach
What sets Omega Optical apart in this field is its consultative approach to problem-solving. Rather than simply providing components, they engage with customers throughout the entire development process. "I think always getting us involved as early as possible in the design is a key aspect," says Washkevich. "The number one thing is getting us involved so that we can help you decide whether or not an existing capability is present, and what the contributing factors to cost and complexity are."
This approach extends beyond just filter design. "We aren't just providing a filter, we're providing a holistic look," he says. "We're not just a filter company now; we're a mirror company, we're a grating company, we do IR optics. We're helping people come up with solutions and understand what results they're going to get."
This approach has led to some remarkable success stories. Washkevich recalls a particularly challenging project: "We had a customer around a year-and-a-half ago show up with a very condensed timeline, and they had decided that they wanted to go from a fixed filter array to a linear variable filter solution on a space-based project."
With a hard deadline dictated by a launch date, Omega not only developed the necessary filters but also coordinated with camera manufacturers and mounting specialists. "We were able to get together four different groups to solve that problem that we were able to sit at the center of and help facilitate," he says. "We actually were able to deliver it two weeks under time and with the data and solutions that they needed."
Future trends
As technology advances, hyperspectral imaging continues to evolve. Washkevich highlights miniaturisation as a key trend: "It is really becoming forefront,” he explains. “Form factors are getting smaller, such as payloads for drones, for example. Everything needs to be smaller and lighter and more cost-efficient from a weight to performance standpoint."
This trend creates new engineering challenges. "Those are really some of the key things that are happening now in the industry, and we are helping to support more dense spectral data over smaller areas," he says.
The company's longevity in the industry, more than 50 years, positions it well to address such evolving needs. Washkevich explains: "Because of our long tenure in the industry, we know a lot of people, and so when new challenges come along, if we don't know how to solve something, we're right there to help you find people that can help us solve it.”
