 We developed a novel approach to automate bird detection tracking in the wild by training a YOLOV4 model on a data set of bird images captured in the wild. The model achieved an average precision of 91.28% on a separate set of test images, outperforming manual identification and tracking methods. This technique has the potential to provide conservationists with more accurate and efficient data about bird populations and their behavior, enabling them to better understand and protect threatened species. This article was authored by Demetrius Mpuziotas, Petrus Carvelis, Ioannis Tsullos, and others.