Automatic Microstructural Classification with Convolutional Neural Network. Lorena, G., Robinson, G., Stefania, P., Pasquale, C., Fabiano, B., & Franco, M. Advances in Intelligent Systems and Computing, pages 170-181. 2019. ![Advances in Intelligent Systems and Computing [link]](https://bibbase.org/img/filetypes/link.svg "link")
Website doi abstract bibtex Microstructural characterization allows knowing the components of a microstructure in order to determine the influence on mechanical properties, such as the maximum load that a body can support before breaking out. In almost all real solutions, microstructures are characterized by human experts, and its automatic identification is still a challenge. In fact, a microstructure typically is a combination of different constituents, also called phases, which produce complex substructures that store information related to origin and formation mode of a material defining all its physical and chemical properties. Convolutional neural networks (CNNs) are a category of deep artificial neural networks that show great success in computer vision applications, such as image and video recognition. In this work we explore and compare four outstanding CNNs architectures with increasing depth to analyze their capability of classifying correctly microstructural images into seven classes. Experiments are done referring to ultrahigh carbon steel microstructural images. As the main result, this paper provides a point-of-view to choose CNN architectures for microstructural image identification considering accuracy, training time, and the number of multiply and accumulate operations performed by convolutional layers. The comparison demonstrates that the addition of two convolutional layers in the LeNet network leads to a higher accuracy without considerably lengthening the training.
@inbook{
type = {inbook},
year = {2019},
keywords = {CNN,Image processing,Microstructure characterization},
pages = {170-181},
websites = {http://link.springer.com/10.1007/978-3-030-02828-2_13},
id = {c9432205-5ed1-3c7d-963b-03b5a43ef939},
created = {2020-12-30T02:11:40.368Z},
file_attached = {false},
profile_id = {e5f1b339-ec56-313b-b123-fd0a1c527f0d},
last_modified = {2020-12-30T02:11:40.368Z},
read = {false},
starred = {false},
authored = {true},
confirmed = {true},
hidden = {false},
citation_key = {Lorena2019},
source_type = {incollection},
private_publication = {false},
abstract = {Microstructural characterization allows knowing the components of a microstructure in order to determine the influence on mechanical properties, such as the maximum load that a body can support before breaking out. In almost all real solutions, microstructures are characterized by human experts, and its automatic identification is still a challenge. In fact, a microstructure typically is a combination of different constituents, also called phases, which produce complex substructures that store information related to origin and formation mode of a material defining all its physical and chemical properties. Convolutional neural networks (CNNs) are a category of deep artificial neural networks that show great success in computer vision applications, such as image and video recognition. In this work we explore and compare four outstanding CNNs architectures with increasing depth to analyze their capability of classifying correctly microstructural images into seven classes. Experiments are done referring to ultrahigh carbon steel microstructural images. As the main result, this paper provides a point-of-view to choose CNN architectures for microstructural image identification considering accuracy, training time, and the number of multiply and accumulate operations performed by convolutional layers. The comparison demonstrates that the addition of two convolutional layers in the LeNet network leads to a higher accuracy without considerably lengthening the training.},
bibtype = {inbook},
author = {Lorena, Guachi and Robinson, Guachi and Stefania, Perri and Pasquale, Corsonello and Fabiano, Bini and Franco, Marinozzi},
doi = {10.1007/978-3-030-02828-2_13},
chapter = {Automatic Microstructural Classification with Convolutional Neural Network},
title = {Advances in Intelligent Systems and Computing}
}
Downloads: 0
{"_id":"JWovaxMKWFbvD377K","bibbaseid":"lorena-robinson-stefania-pasquale-fabiano-franco-advancesinintelligentsystemsandcomputing-2019","author_short":["Lorena, G.","Robinson, G.","Stefania, P.","Pasquale, C.","Fabiano, B.","Franco, M."],"bibdata":{"type":"inbook","year":"2019","keywords":"CNN,Image processing,Microstructure characterization","pages":"170-181","websites":"http://link.springer.com/10.1007/978-3-030-02828-2_13","id":"c9432205-5ed1-3c7d-963b-03b5a43ef939","created":"2020-12-30T02:11:40.368Z","file_attached":false,"profile_id":"e5f1b339-ec56-313b-b123-fd0a1c527f0d","last_modified":"2020-12-30T02:11:40.368Z","read":false,"starred":false,"authored":"true","confirmed":"true","hidden":false,"citation_key":"Lorena2019","source_type":"incollection","private_publication":false,"abstract":"Microstructural characterization allows knowing the components of a microstructure in order to determine the influence on mechanical properties, such as the maximum load that a body can support before breaking out. In almost all real solutions, microstructures are characterized by human experts, and its automatic identification is still a challenge. In fact, a microstructure typically is a combination of different constituents, also called phases, which produce complex substructures that store information related to origin and formation mode of a material defining all its physical and chemical properties. Convolutional neural networks (CNNs) are a category of deep artificial neural networks that show great success in computer vision applications, such as image and video recognition. In this work we explore and compare four outstanding CNNs architectures with increasing depth to analyze their capability of classifying correctly microstructural images into seven classes. Experiments are done referring to ultrahigh carbon steel microstructural images. As the main result, this paper provides a point-of-view to choose CNN architectures for microstructural image identification considering accuracy, training time, and the number of multiply and accumulate operations performed by convolutional layers. The comparison demonstrates that the addition of two convolutional layers in the LeNet network leads to a higher accuracy without considerably lengthening the training.","bibtype":"inbook","author":"Lorena, Guachi and Robinson, Guachi and Stefania, Perri and Pasquale, Corsonello and Fabiano, Bini and Franco, Marinozzi","doi":"10.1007/978-3-030-02828-2_13","chapter":"Automatic Microstructural Classification with Convolutional Neural Network","title":"Advances in Intelligent Systems and Computing","bibtex":"@inbook{\n type = {inbook},\n year = {2019},\n keywords = {CNN,Image processing,Microstructure characterization},\n pages = {170-181},\n websites = {http://link.springer.com/10.1007/978-3-030-02828-2_13},\n id = {c9432205-5ed1-3c7d-963b-03b5a43ef939},\n created = {2020-12-30T02:11:40.368Z},\n file_attached = {false},\n profile_id = {e5f1b339-ec56-313b-b123-fd0a1c527f0d},\n last_modified = {2020-12-30T02:11:40.368Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {true},\n hidden = {false},\n citation_key = {Lorena2019},\n source_type = {incollection},\n private_publication = {false},\n abstract = {Microstructural characterization allows knowing the components of a microstructure in order to determine the influence on mechanical properties, such as the maximum load that a body can support before breaking out. In almost all real solutions, microstructures are characterized by human experts, and its automatic identification is still a challenge. In fact, a microstructure typically is a combination of different constituents, also called phases, which produce complex substructures that store information related to origin and formation mode of a material defining all its physical and chemical properties. Convolutional neural networks (CNNs) are a category of deep artificial neural networks that show great success in computer vision applications, such as image and video recognition. In this work we explore and compare four outstanding CNNs architectures with increasing depth to analyze their capability of classifying correctly microstructural images into seven classes. Experiments are done referring to ultrahigh carbon steel microstructural images. As the main result, this paper provides a point-of-view to choose CNN architectures for microstructural image identification considering accuracy, training time, and the number of multiply and accumulate operations performed by convolutional layers. The comparison demonstrates that the addition of two convolutional layers in the LeNet network leads to a higher accuracy without considerably lengthening the training.},\n bibtype = {inbook},\n author = {Lorena, Guachi and Robinson, Guachi and Stefania, Perri and Pasquale, Corsonello and Fabiano, Bini and Franco, Marinozzi},\n doi = {10.1007/978-3-030-02828-2_13},\n chapter = {Automatic Microstructural Classification with Convolutional Neural Network},\n title = {Advances in Intelligent Systems and Computing}\n}","author_short":["Lorena, G.","Robinson, G.","Stefania, P.","Pasquale, C.","Fabiano, B.","Franco, M."],"urls":{"Website":"http://link.springer.com/10.1007/978-3-030-02828-2_13"},"biburl":"https://bibbase.org/service/mendeley/e5f1b339-ec56-313b-b123-fd0a1c527f0d","bibbaseid":"lorena-robinson-stefania-pasquale-fabiano-franco-advancesinintelligentsystemsandcomputing-2019","role":"author","keyword":["CNN","Image processing","Microstructure characterization"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inbook","biburl":"https://bibbase.org/service/mendeley/e5f1b339-ec56-313b-b123-fd0a1c527f0d","dataSources":["2252seNhipfTmjEBQ"],"keywords":["cnn","image processing","microstructure characterization"],"search_terms":["advances","intelligent","systems","computing","lorena","robinson","stefania","pasquale","fabiano","franco"],"title":"Advances in Intelligent Systems and Computing","year":2019}