Implementation of an Algorithm for Automated Bird Counting in the Poultry Sector

León León R.A.; Negreiros Alonzo H.Y.; Aguilar Artiaga D.; Ponce León C.C.; Salinas Polo E.; Chávez Trelles F.A. · Memorias de la Conferencia Iberoamericana de Complejidad, Informatica y Cibernetica, CICIC (2025) · DOI: 10.54808/cicic2025.01.299

Resumen

The poultry industry faces challenges in production control of bird counting. Artificial vision algorithms improve operational efficiency and animal welfare. These systems allow automated bird counting, reducing time. Python is key in its development, also facilitating the creation of interfaces and the optimization of algorithms. This experimental research collected data using a 720px Webcam and a Core i3 laptop. YOLOv8 was used to create the neural structure by capturing images of the birds, to perform image processing, the neural network was trained, evaluated, the system was implemented and concluded with machine learning. The results demonstrated an accuracy of 85.36% when detecting and counting birds. In conclusion, this technology is reliable and effective, with accurate bird counting, helping in the poultry industry process. Copyright 2025. © por el International Institute of Informatics and Systemics.

Instituciones: Universidad Privada del Norte