Artificial Vision System for the Classification of Green Lemons (Citrus Latifolia)

Technology ceased to be an exclusivity in the world, its current use is necessary and fundamental in the industrial field. In such a competitive and globalized world, technology optimizes the production processes of organizations. In the agri-food sector, exporting companies need to improve their processes of categorization and homogenization of their products based on the uniformity of its physical characteristics, such is the case of the Peruvian green lemon. Therefore, it is necessary to save time, cost and human effort in carrying out this process. Therefore, this research aims to implement an artificial vision system for the classification of the dimension of lemons. The artificial vision system consists of the processes of image capture, pre-processing, segmentation, description and recognition and interpretation. Where algorithms compare the characteristics of the lemon to obtain an optimal result. As a result, after using the artificial vision system, an efficiency of 85.1%, a sensitivity of 84% and a specificity of 100% were obtained. It is concluded that the detection of the caliber of the lemons works optimally allowing the number of hits to meet the objective of the investigation. The implementation of this system provides a solution to the problems of product homogeneity.