Hyperspectral imaging (HSI) can provide valuable information about the spatial distribution of ingredients in an object, therefore the technique has been widely adopted in numerous applications, ranging from remote sensing and land planning, food quality control, to biomedical applications. However, HSI instruments are expensive, which has limited the technique to some high-end applications. In this study, we developed a cost-effective HSI technique with an internal scene-scan mechanism, which enables rapid acquisitions of a scene without moving the instrument or the tested object. The apparatus was characterised, revealing an imaging resolution of 0.4 mm in a field of view (FoV) of 10 cm and a spectral resolution of 1.3 nm in the 40–800 nm visible light region. We succeeded in applying our apparatus to analyse the oxidation processes of apple and meat, which demonstrated our design and relevant data analysis to be of high value to visualise chemistry related to food quality and safety.
Keywords: hyperspectral imaging, diffuse reflectance, visible light, food quality control, telecentric, principal component analysis, L2 norm, spectral angle, k-mean clustering