One of the main issues related to the treatment of soybean seeds concerns its vigor definition. The great majority of vigor analysis is performed by a human specialist leading to an extremely tiresome and subjective approach, highly susceptible to failures. In order to deal with this problem the present paper proposed a methodology joining content-based image retrieval techniques with the specialist perception to retrieve soybean seeds according to his/her expectation and aiding the seed vigor definition. The experiments performed showed that the proposed approach presented several notable contributions to the soybean vigor definition process, reaching up to 100% of precision regarding the similarity queries. Moreover, it not only improves in a great extent the entire process from planting to harvesting, but also enables to automate and accelerate it.