Brassica is a genus of oilseed plants mainly used to produce edible oils, modified lipids, industrial oils, and biofuels. Oil and fatty acid content are the main chemical indicators for Brassicas seed quality (e.g. low content of erucic acid indicate seeds appropriate for food industry, while high contents indicate are suitable in the cosmetic, pharmaceutical and fuel industry). The goal of this work was to implement and compare the portable Near Infrared spectroscopy (NIRS) and NIR-Hyperspectral Imaging (NIR-HSI) based analytical methods to quantify oil content and fatty acid and classify seeds species. Spectral data was analyzed by non-supervised (principal component analysis, PCA) and supervised (partial least square regression, PLSR, and discriminant analysis, PLS-DA) chemometrics tools in order to generate new prediction models. PLS-DA analysis showed satisfactory discrimination between Brassicas species, with correct classification rate of 94.9 and 100 % for portable NIR spectrometer and NIR-HSI devices, respectively, in external validation. The best prediction models were obtained based on interval selection (iPLS) for erucic acid, MUFAs and PUFAs using NIR-HSI spectra. Although these NIR-HSI models have better results than the NIR spectrometer, both the NIR and NIR-HSI devices could be adapted to quantify the oil content and composition in Brassica seeds, according to the needs of the industry or the consumer.