Resumen

FLORES COLORADO, Oscar Eder; CERVANTES CANALES, Jair; GARCIA-LAMONT, Farid  y  RUIZ CASTILLA, José Sergio. Identification of the main diseases of the coffee plant (Coffea arabica) through artificial vision. Cienc. ergo-sum [online]. 2023, vol.30, n.3, e212.  Epub 14-Ene-2025. ISSN 2395-8782.  https://doi.org/10.30878/ces.v30n3a8.

Pattern recognition techniques in digital images are applied to identify healthy leaves and four important diseases of the coffee plant Coffeea arabica. The diseases are coffee rust, leaf miner, Phoma leaf spot and Cercospora coffeicola. To achieve this, different segmentation techniques were used, among them: Otsu, PCA and Global Border Method. To get the features vector, the images were processed to extract the Chromatic, geometric and textural features. Finally, four classification algorithms were implemented, including support vector machine, random forest, Naive Bayes and Artificial neural networks backpropagation. The best accuracy obtained is 83% with Otsu segmentation and classification with backpropagation Artificial neural networks.

Palabras llave : pattern recognition; artificial vision; coffee plant diseases.

        · resumen en Español     · texto en Español     · Español ( pdf )