Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases

This project presents an iterative approach for upscaling a machine learning model for microstructural semantic segmentation of two-phase steels light optical micrographs. Several deep learning models have been trained, using a U-NET architecture with DenseNet-201 pretrained weights on ImageNet as b...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Gonzalez Santucho, Lautaro Elbio
Otros Autores: Moran, Juan Ignacio
Formato: Tesis acceptedVersion Tesis de grado
Lenguaje:Inglés
Publicado: Universidad Nacional de Mar del Plata. Facultad de Ingeniería. Argentina 2024
Materias:
Machine learning
Aceros
Aceros de dos fases
Acceso en línea:http://rinfi.fi.mdp.edu.ar/handle/123456789/910
Aporte de:
RINFI - Facultad de Ingeniería (UNMdP) de Universidad Nacional de Mar del Plata (UNMdP)
id I29-R182-123456789-910
record_format dspace
spelling I29-R182-123456789-9102024-08-27T16:12:07Z Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases Gonzalez Santucho, Lautaro Elbio Moran, Juan Ignacio Bachmann, Björn Ivo Machine learning Aceros Aceros de dos fases This project presents an iterative approach for upscaling a machine learning model for microstructural semantic segmentation of two-phase steels light optical micrographs. Several deep learning models have been trained, using a U-NET architecture with DenseNet-201 pretrained weights on ImageNet as backbone. Metallographic samples from rolled plates have been produced and analyzed in different microscopes to collect data for training and testing, aiming to specifically increase the manageable variance as well as the model’s robustness. The results from previous models were then used as masks to train the final one. The incorporation of a higher variance in the model through different acquisition conditions images resulted in a more robust model, that can consistently segment images at various magnifications, from different microscopes, and taken under suboptimal conditions. The utilization of previous segmentation results as masks allowed to introduce more data to the training data set. This allowed to minimize the need to produce hand annotated masks, which are time consuming to make and often constitute a bottle neck for model training. The relevance of these results lies in the possibility to correlate the results from the model (second phase fraction and morphological parameters of the particles) with mechanical properties and manufacturing parameters. Moreover, light optical micrographs are inexpensive, fast to produce and already implemented in quality control at an industrial scale, thus making the implementation of this analysis technique in the industry feasible. Fil: Gonzalez Santucho, Lautaro Elbio. Universidad Nacional de Mar del Plata. Facultad de Ingeniería; Argentina 2024-08-09 Thesis info:eu-repo/semantics/acceptedVersion info:ar-repo/semantics/tesis de grado info:eu-repo/semantics/bachelorThesis application/pdf http://rinfi.fi.mdp.edu.ar/handle/123456789/910 eng info:eu-repo/semantics/openAccess https://creativecommons.org/licenses/by/4.0/ Universidad Nacional de Mar del Plata. Facultad de Ingeniería. Argentina
institution Universidad Nacional de Mar del Plata (UNMdP)
institution_str I-29
repository_str R-182
collection RINFI - Facultad de Ingeniería (UNMdP)
language Inglés
topic Machine learning
Aceros
Aceros de dos fases
spellingShingle Machine learning
Aceros
Aceros de dos fases
Gonzalez Santucho, Lautaro Elbio
Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
topic_facet Machine learning
Aceros
Aceros de dos fases
description This project presents an iterative approach for upscaling a machine learning model for microstructural semantic segmentation of two-phase steels light optical micrographs. Several deep learning models have been trained, using a U-NET architecture with DenseNet-201 pretrained weights on ImageNet as backbone. Metallographic samples from rolled plates have been produced and analyzed in different microscopes to collect data for training and testing, aiming to specifically increase the manageable variance as well as the model’s robustness. The results from previous models were then used as masks to train the final one. The incorporation of a higher variance in the model through different acquisition conditions images resulted in a more robust model, that can consistently segment images at various magnifications, from different microscopes, and taken under suboptimal conditions. The utilization of previous segmentation results as masks allowed to introduce more data to the training data set. This allowed to minimize the need to produce hand annotated masks, which are time consuming to make and often constitute a bottle neck for model training. The relevance of these results lies in the possibility to correlate the results from the model (second phase fraction and morphological parameters of the particles) with mechanical properties and manufacturing parameters. Moreover, light optical micrographs are inexpensive, fast to produce and already implemented in quality control at an industrial scale, thus making the implementation of this analysis technique in the industry feasible.
author2 Moran, Juan Ignacio
author_facet Moran, Juan Ignacio
Gonzalez Santucho, Lautaro Elbio
format Thesis
acceptedVersion
Tesis de grado
Tesis de grado
author Gonzalez Santucho, Lautaro Elbio
author_sort Gonzalez Santucho, Lautaro Elbio
title Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
title_short Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
title_full Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
title_fullStr Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
title_full_unstemmed Clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
title_sort clasificación y cuantificación basada en aprendizaje automático de aceros de dos fases
publisher Universidad Nacional de Mar del Plata. Facultad de Ingeniería. Argentina
publishDate 2024
url http://rinfi.fi.mdp.edu.ar/handle/123456789/910
work_keys_str_mv AT gonzalezsantucholautaroelbio clasificacionycuantificacionbasadaenaprendizajeautomaticodeacerosdedosfases
_version_ 1809230053360271360

Ejemplares similares