BATISTA, G. A.; http://lattes.cnpq.br/2950618164501266; BATISTA, Gabriel de Azevedo.
Résumé:
Cowpea is a crop of great importance worldwide, which is why many native varieties and improved cultivars are explored. The seeds of these materials have intrinsic characteristics that distinguish them. Thus, the aim of this work was to adjust the machine learning models for cowpea identification based on the processing of digital images of seeds using artificial intelligence techniques. The research is of the qualitative and quantitative type, and was carried out in the teachers' environment, room 03. Therefore, digital images of seeds of 10 (ten) cultivars were obtained and processed using the vectorizers InceptionV3, SqueezeNet, VGG16 and VGG19. Subsequently, the learning algorithms of K-Nearest Neighbors (KNN - number of nearest neighbors), Decision Tree (Tree), Random Forest (RF), Gradient Boosting (GB), Vector Support Machine (SVM - Support Vector Machines) and Artificial Neural Network (MLP - Multi-Layer Perceptron). The machine learning algorithms Artificial Neural Network and Vector Support Machine (SVM) had better performance indicators for cowpea identification from the processing of digital images of seeds. This work contributes, both in the academic field and in the practical aspect, considering that the available data can serve in the future for the creation of mobile applications.