Design of a Convolutional Network for Recognition of birds at the Universidad Libre de Bogotá Sede el Bosque

Abstract

Accelerated urban growth and the expansion of metropolitan areas pose challenges for biodiversity conservation, especially in species-rich regions. As cities grow, natural habitats become fragmented and degraded, which can lead to species declines.

Academic activity and the constant presence of students on university campuses also generate disturbances in the natural habitat. The construction of infrastructure and human traffic can alter the behavior of species. It is crucial to find methods to monitor and mitigate these impacts, contributing to conservation in academic and urban environments.

This study proposes to design a prototype of an artificial neural network to analyze the behavior of two species of birds at the Universidad Libre de Bogotá, El Bosque headquarters. These networks, inspired by the human brain, offer a powerful approach to analyzing complex data, identifying behavioral patterns valuable for conservation.

The methodology, based on the Scrum framework, and the preliminary results are presented, which have been satisfactory, managing to accurately identify the birds studied. These findings demonstrate the potential of neural networks for biodiversity conservation in urban environments, providing accurate data to design effective conservation strategies.