Varietal quality control in the nursery plant industry using computer vision and deep learning techniques

Abstract

Computer vision coupled to deep learning is a promising technique withmultiple applications in the industry. In this work, the potential of thistechnique has been assessed in the classification of two varieties of almondtrees (Prunus dulcis), Soleta and Pentacebas. For that, a convolutional neuralnetwork named VGG16 was used. The most appropriate configuration formodel training was studied, which included the comparison between twodifferent filling modes (reflect and nearest) in the data augmentation step, theevaluation of the batch size and the analysis of the image sizes. The robustnessof the model was also checked, and information was obtained about how themodel extracts the information from the images