Optical spectroscopy provides straightforward solutions to many requirements of quality and safety controls for the food and beverage industry.It enables a rapid and non-destructive analysis without making use of reagents or chemical treatments, thus avoiding the problem of waste discharge.Since it does not imply environmental side effects, it is an ideal candidate for "green analytics".The VIS spectrum reveals the presence of dyes and pigments, typically, chlorophylls, polyphenols, and carotenoids.The NIR spectrum is informative for proteins, fats and carbohydrates, thanks to the presence of overtones and combinations of vibrational modes of C-H, O-H, and N-H bonds.Indeed, this cumulative VIS-NIR spectrum can be considered a sort of optical signature of the food, the univocal fingerprint from which to extract the analysis of several constituents, as well as the evaluation of quality indicators, or the authenticity of the product.Spectroscopy-based sensors making use of micro-optic and miniaturized components were capable to offer effective and low-cost solutions for many industrial and process control applications.Especially for the food sector, the intrinsic optical and mechanical characteristics of optical fibers, together with the wide availability of bright LEDs and portable spectrometers, made it possible to implement compact instrumentation with a high potential for many applications.Given the nature and complexity of the spectroscopic data involved, multivariate and chemometric processing techniques are needed to provide qualitative information for product authentication, or more sophisticated predictive models for quantifying quality and safety indicators.Indeed, smart data processing made VIS-NIR optical spectroscopy a rapid and non-destructive method for the multicomponent analysis of foodstuffs.Since water has a weak absorption in these spectral bands, also high-moisture foods were easily analyzed.This paper shows how absorption spectroscopy combined with chemometrics demonstrated effectiveness for detecting quality indicators of extra virgin olive oils, and for highlighting counterfeit using lower quality fats both in olive oils and soybean oils.Conventional absorption spectroscopy in the entire 400-2500 nm band was used for detecting the region of origin and the most important quality indicators (peroxides and fatty acids) of a set of Italian extra virgin olive oils.An innovative setup for diffuse light absorption spectroscopy in the 400-1700 nm band was experimented to detect the fraud of high quality extra virgin olive oils caused by lower quality olive oils.A simple setup for absorption spectroscopy in the visible band only was used to detect Chinese soy bean oils mixed with swill oil dredged from gutters behind restaurants.Also, perspectives will be given for quality detection of other popular foodstuffs, such as honey, milk, beer,tomatoes, and others.