Torrecilla Velasco, José SantiagoPradana Lopez, SandraPérez Calabuig, Ana M.Cancilla, John C.Lozano, Miguel AngelRodrigo, CarlosMena, Maria Luz2026-01-122026-01-122021-04-01Pradana-López, Sandra, et al. «Deep Transfer Learning to Verify Quality and Safety of Ground Coffee». Food Control, vol. 122, abril de 2021, p. 107801. DOI.org (Crossref), https://doi.org/10.1016/j.foodcont.2020.107801.10.1016/j.foodcont.2020.107801https://hdl.handle.net/20.500.14352/129844This record corresponds to the peer-reviewed journal article “Deep transfer learning to verify quality and safety of ground coffee”, published in Food Control (Volume 122, 2021), a Journal Citation Reports (JCR) indexed journal. The work presents the development and validation of a computer vision–based methodology for coffee quality control and fraud detection, using convolutional neural networks (CNNs) combined with transfer learning. Visible-light images of ground coffee samples were acquired with a standard photographic camera and processed using a ResNet34 architecture, enabling the classification of Arabica and Robusta coffees as well as the detection and quantification of adulterations with chicory and barley. The proposed models achieved high classification performance, with overall accuracies above 98% and the capability to detect adulterant contents as low as 0.5% (w/w) across different particle size ranges. The results demonstrate the robustness and applicability of deep transfer learning approaches for rapid, low-cost food authentication and safety assessment, with potential relevance for producers, distributors, and consumers.engjournal articlehttps://doi.org/10.1016/j.foodcont.2020.107801restricted access66.0Coffee classificationFood qualityAdulterationPhotographic cameraConvolutional neural networksTransfer learningResNet34Ingeniería química3309 Tecnología de Los Alimentos