Multivariate data modelling for de-shadowing of airborne hyperspectral imaging

Abstract

Airborne hyperspectral imaging is a powerful technique for high-resolution classification of large areas of ground, applied today in fields like agriculture and environmental monitoring. Even though many classification algorithms are capable of handling shadows without a decrease in performance, visual inspection can be made easier if shadows are removed. In this paper we present a method for separating the effect of shadows (de-shadowing) and other partially known lighting condition changes from the effects due to the physical, chemical or biological properties of the ground, which are of interest. An example application is shown with good results.

Classification in hyperspectral images by independent component analysis, segmented cross-validation and uncertainty estimates

“…Prior to modelling of the hyperspectral data, preprocessing the images in the image domain, but foremost in the spectral domain, is usually needed (e.g., background removal, scatter correc-tion, de-noising, suppression of sample morphology effects or treatment of dead pixels). 1,8 These big data, structured as hyperspectral image cubes, have relevance in many types of applications, for example agricultural and food sciences, 9,10 for data collection by drones 11 and in the pharmaceutical industry. 12 The applicability of multivariate data analysis for HSI is relevant for process analytical control (PAC) and quality by design (QbD) in a wide range of industrial sectors.…”