In common beans, color is a physical property related to nutritional compounds that are beneficial to health; in this work, we explored the quantification of color to know the different proportions of color in samples of common bean landraces. This research is significant as it sheds light on the importance of color quantification in nutritional analysis. A dataset was created with colorimetric information of seeds of seven different colorations that was used for the k-nearest neighbors algorithm for assigned labels; 168 images of bean landraces of heterogeneous color were processed to assign labels to each one; the labels generated for the seeds of each bean landrace allowed us to quantify the color; 121 samples the 100% of grain were correctly labeled, on the other hands, in 47 bean samples, the labels assigned to some seeds did not describe to the color of the seed. The results show that it is possible to quantify the color by classifying the set of representative seeds of bean landraces.