Application

Spectral Analytix and Resonon

By Dr. Rand Swanson and Dr. Christian Nansen (Spectral Analytix) - February 8, 2024

Hyperspectral imaging can be used in nearly any application where one wishes to distinguish between two or more classes of objects, such as: viable and non-viable seeds, contaminated and uncontaminated water, types of paper, coffee roasting classes, and much more. It would be impossible to obtain expertise across such a broad range of applications, so Resonon directly supports companies that have the necessary expertise by helping them implement hyperspectral solutions. One of those companies is Spectral Analytix.

After having used Resonon’s hyperspectral cameras for over 20 years in academia, Christian Nansen founded Spectral Analytix.

Figure 1: Seeds traveling on conveyor belt of Spectral Analytix machine vision system.

Spectral Analytix has developed an advanced machine vision system with a target market of the seed industry, detailed in the video below. The hyperspectral sensor can detect very detailed reflectance features, which are used to classify and sort germinating from non-germinating seeds, and to determine germ size, ploidy (the number of sets of chromosomes), sex of seeds, and genotypic variation. The end result is that seed companies can optimize crop breeding programs by saving time and resources, and ultimately deliver more high-quality seeds to growers.

As a university professor, Christian has published extensively on specific applications of hyperspectral imaging to classify objects [1-6], about ways to examine performance of classification functions [1] and about how to maintain radiometric repeatability of hyperspectral imaging data [7].

Christian puts it this way: “As researchers, we have learned how powerful hyperspectral imaging is – that it can be used to detect very subtle differences between classes of objects, such as germinating and non-germinating seeds. This is the exciting part and what is demonstrated in a wealth of research articles. But, we have also learned that repeatability of classifications can be tricky… that multiple factors affect the consistency and quality of hyperspectral imaging data collected at different time points. Knowledge about and control of these variables are essential to the development of reliable and accurate solutions for our clients.”

Figure 2: Christian Nansen reviewing hyperspectral data.

By using their system to provide a service, Spectral Analytix meets client needs without the client needing to purchase hardware and learn how to analyze hyperspectral data. Resonon is excited to see Spectral Analytix achieve commercial success and we look forward to working together for years to come.

If you have an application that might benefit from the insights that hyperspectral imaging can provide, let us know. Our Sales team is happy to discuss your application and will run sample scans to help you see what hyperspectral imaging can do for you.

References

1. Experimental data manipulations to assess performance of hyperspectral classification models of crop seeds and other objects.

Nansen, C.; Imtiaz, M.S.; Mesgaran, M.B.; Lee, H. Plant Methods 2022, 18, 74, doi:10.1186/s13007-022-00912-z. LINK

2. Early infestations by arthropod pests induce unique changes in plant compositional traits and leaf reflectance.

Nansen, C.; Murdock, M.; Purington, R.; Marshall, S. Pest Management Science 2021, doi:https://doi.org/10.1002/ps.6556. LINK

3. Hyperspectral remote sensing to detect leafminer-induced stress in bok choy and spinach according to fertilizer regime and timing.

Nguyen, H.; Nansen, C. Pest Management Science 2020, 76, 2208-2216, doi:10.1002/ps.5758. LINK

4. Proximal remote sensing to differentiate nonviruliferous and viruliferous insect vectors – proof of concept and importance of input data robustness.

Nansen, C.; Stewart, A.N.; Gutierrez, T.A.M.; Wintermantel, W.M.; McRoberts, N.; Gilbertson, R.L. Plant Pathology 2019, 68, 746-754, doi:https://doi.org/10.1111/ppa.12984. LINK

5. Using hyperspectral imaging to characterize consistency of coffee brands and their respective roasting classes.

Nansen, C.; Singh, K.; Mian, A.; Allison, B.J.; Simmons, C.W. Journal of Food Engineering 2016, 190, 34-39, doi:http://dx.doi.org/10.1016/j.jfoodeng.2016.06.010. LINK

6. Using hyperspectral imaging to determine germination of native Australian plant seeds.

Nansen, C.; Zhao, G.; Dakin, N.; Zhao, C.; Turner, S.R. Journal of Photochemistry and Photobiology B: Biology 2015, 145, 19-24, doi:10.1016/j.jphotobiol.2015.02.015. LINK

7. Calibration to maximize temporal radiometric repeatability of airborne hyperspectral imaging data.

Nansen, C.; Lee, H.; Mantri, A. Frontiers in Plant Science 2023, 14. LINK