IMC have collaborated with both universities and industry to utilise spectral imaging to solve otherwise unsolvable problems.
What is spectral imaging?
The human eye sees visible light in three bands (red, green, and blue), this is similar to how standard cameras capture light.
Spectral imaging works differently by dividing the spectrum into many more bands - sometimes hundreds. It preserves the scenes spectral content, along with much hidden information, by measuring radiance at each of these narrow bands. By looking at the spectral signature we are able to view information about the chemical composition or the physical structure of an object that is otherwise invisible. For example, it may let us assess the health of vegetation or sort one type of plastic from another.
Isn't hyperspectral imaging expensive?
Not always! IMC have an in-house hyperspectral imaging camera, along with numerous high-end imagers (covering UV, VIS, VNIR, NIR, SWIR, MWIR, LWIR). These cameras are often prohibitively expensive for many applications. However, often they are not required. A typical workflow for the development of a spectral imaging system involves:
- Data is captured using the best imaging sensors and detectors available. IMC can assist with ensuring that data is collected in a controlled and consistent manner, to provide radiometrically calibrated results across a broad spectral range.
- Our engineers will use our purpose developed analysis tools to assess the captured data (spectral data "cubes"). The aim of this analysis is to identify the key spectral information that is required for discrimination and classification in your specific application.
- We provide a multispectral system design that uses sensors to assess only the spectral information that is relevant for the application. This may be a very low cost camera with a low cost optical filter set - an efficient design only made possible by starting with the hyperspectral dataset.