Sistema de visión artificial para el monitoreo de pollos de engorde en la granja avícola Mascriollo.SAS de la ciudad de Cúcuta.

Abstract

In the process of production of broiler, stress is a critical factor that influences the final quality of the product. In this work an image acquisition system was developed and implemented for the analysis and detection of behaviors associated with stress situations in fattening birds as a control indicator. This indicator allows generating early warnings against alterations in the behavior of birds that have a negative impact on the welfare of animals. In the processing of images, some segmentation techniques were applied (binarization by thresholding, binarization by Otsu method, dilation) that allowed based on extraction of centroids to estimate the number of birds present in a certain segment and their activity patterns. Those who performed the processing of the images was set in ten seconds in order to observe the movements of the birds. Finally, it was demonstrated that artificial vision and image processing make an important tool for the analysis of fattening birds.