Abstract:

Machines and software represent technology in the production processes of exporting companies. The competition in the food products market is for technological development and the quality of its products to satisfy the need of consumers for increasingly demanding products. Automation is essential to streamline general product controls and minimum quality requirements in tuber classification. This research proposes as its main objective to put into operation a computer vision system for the classification of the black potato solanun chaucha. The system is developed with the potato classification tests starting with the digital imaging technique process, followed by the image processing stages; capture, pre-processing, segmentation, description and recognition of images. Obtaining as a result 78.2% of true positives VP, 14.6% of true negatives VN, 0% of false positives FP and 7.2% of false negatives FN. According to the results, the classification time of 0.44 seconds demonstrates the effectiveness of the system. From the data. We can conclude that it is feasible and viable to carry out the classification of black potatoes through the computer vision system.