In this article, digital image processing and analysis (DIPA) combined with chemometric methods, principal component analysis (PCA) and hierarchical cluster analysis (HCA) were used to discriminate sesame seeds through their digitized images. For this purpose, four groups of seeds were used: BRS Anahí and BRS Seda cultivars, a lineage and a commercial sample. The images were scanned using an HP officejet 7610 scanner and, for extraction of the red-green-blue channels and colorimetric profile, the ImageJ software was used. The DIPA combined with chemometric methods allowed us to discriminate the four groups of sesame seeds efficiently, and a minimum accumulated variance of 89.03% of the total variance was obtained. The trends observed via the PCA were confirmed through the dendrograms obtained using the HCA. The results achieved in this work indicate that the proposed methodology can be a simple analytical alternative for the non-destructive phenotypic discrimination of seeds, with their color as an attribute.