Hyperspectral image analysis works with image data where each pixel contains hundreds of wavelengths acquired from spectral measurements. It is a growing field of research in the sciences and industries because it can distinguish visually similar objects. While many machine-learning methods work well for analysing regular images, little is known about how they perform on hyperspectral data. Standard methods for quantifying and classifying hyperspectral data include the chemometric methods PLS, PLS-DA and SIMCA. They provide rapid computations along with intuitive modelling and diagnostic tools, but cannot capture more complex data. I benchmarked the chemometric methods against machine learning methods from Microsoft's ML.NET library on six classification and two quantification problems. The ML.NET methods proved to be good complements to the chemometric methods. In particular, the decision tree methods provided accurate classification and quantification while the maximum entropy classification methods balanced between accuracy and computational time the best. While the remaining ML.NET methods performed equally well or better than the chemometric methods, finding their use requires testing on data sets with a wider range of properties. The best ML.NET methods are suitable for analysing more complex hyperspectral images by capturing nonlinearities disregarded by standard image analysis.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:umu-184538 |
| Date | January 2021 |
| Creators | Vikström, Axel |
| Publisher | Umeå universitet, Institutionen för fysik |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.011 seconds