RT info:eu-repo/semantics/article T1 On the use of neural networks for the structural characterization of polymeric porous materials A1 Torre Ordás, Jorge A1 Barroso Solares, Suset A1 Rodríguez Pérez, Miguel Ángel A1 Pinto Sanz, Javier K1 Polimeros y polimerización K1 Deep learning K1 Mask R–CNN K1 Automatic K1 Aprendizaje profundo K1 Automático K1 Máscara R-CNN K1 2301.15 Análisis de Polímeros AB The structural characterization is an essential task in the study of porous materials. To achieve reliable results, it requires to evaluate images with hundreds of pores. Current methods require large time amounts and are subjected to human errors and subjectivity. A completely automatic tool would not only speed up the process but also enhance its reliability and reproducibility. Therefore, the main objective of this article is the study of a deep-learning-based technique for the structural characterization of porous materials, through the use of a convolutional neural network. Several fine-tuned Mask R–CNN models are evaluated using different training configurations in four separate datasets each composed of numerous SEM images of diverse polymeric porous materials: closed-pore extruded polystyrene (XPS), polyurethane (PU), and poly(methyl methacrylate) (PMMA), and open-pore PU. Results prove the tool capable of providing very accurate results, equivalent to those achieved by time-consuming manual methods, in a matter of seconds. PB Elsevier SN 0032-3861 YR 2024 FD 2024 LK https://uvadoc.uva.es/handle/10324/66062 UL https://uvadoc.uva.es/handle/10324/66062 LA eng NO Polymer, 2024, vol. 291, 126597 NO Producción Científica DS UVaDOC RD 12-jul-2024