Artificial Vision System for the inspection of dates in industrial food products (#526)

Authors

León León, Ryan Abraham

García Saire, Jair Milton

Abstract

The main objective of this work is to improve the efficiency in the inspection of dates in industrial food products through an artificial vision system using the Python language, which will make it possible to process the images for their subsequent classification between an image of correct or incorrect visualization, contributing to future studies and development of similar systems. It began by identifying the main problems that exist in most of the food industries worldwide. Likewise, based on the conclusions found in previously developed works. Once the theoretical base was obtained, we proceeded to identify the necessary materials to carry out this project and carry out the processing where the correct visualization of the expiration dates of said products is verified. At the end of the control, the approved products go to the dispatch area for marketing, otherwise, it will be rejected as a defective product. Various tests were carried out that conclude that the artificial vision system together with the Python language shows an efficiency of 97.20% reducing human error, proving its effectiveness in the quality process.