Identification of Solid Waste Types to Optimize the Recycling Process Through Deep Learning Techniques

The inadequate management of solid waste constitutes a considerable environmental challenge, causing pollution and waste of resources that negatively impact the sustainability of the planet. In order to address this challenge, the present research focuses on automatic waste classification using Deep Learning (DL) techniques. For this purpose, the TrashNet dataset was employed and three different Convolutional Neural Networks (CNNs) architectures were tested: AlexNet, YOLOv8 (You Only Look Once) and YOLOv11. The objective of this article is twofold: firstly, to enhance the accuracy of waste identification; and secondly, to achieve high levels of efficiency in its classification. In doing so, it contributes to the development of technological solutions that optimize the recycling process, as well as promoting a more sustainable management of materials.