Method for classifying phases of metallographic structure, device for classifying phases of metallographic structure, method for learning phases of metallographic structure, device for learning phases of metallographic structure, method for predicting material properties of metal material, and device for predicting material properties of metal material

This method for classifying phases of a metallographic structure comprises: a feature value calculation step for calculating one or more feature values for pixels of an image obtained by imaging a metallographic structure of a metal material; and a phase classification step in which the feature valu...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: KIYOKANE NAOYA, OBATA YOSHIE, YAMASHITA TAKAKO
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This method for classifying phases of a metallographic structure comprises: a feature value calculation step for calculating one or more feature values for pixels of an image obtained by imaging a metallographic structure of a metal material; and a phase classification step in which the feature values calculated in the feature value calculation step are inputted to a learning completion model in which learning is performed using the feature values of the labels of the plurality of phases of the metallographic structure as inputs and the labels of the respective phases as outputs, and the labels of the phases of the pixels corresponding to the inputted feature values are obtained. The phases of the metallographic structure of the image are thereby classified. 金相组织的相的分类方法包括:特征值计算步骤,其中,对拍摄金属材料的金相组织而得到的图像的像素计算一个以上的特征值;和相分类步骤,其中,对于以被赋予了金相组织的多个相的标签的特征值作为输入、以各相的标签作为输出进行学习后的学习完成模型,输入在特征值计算步骤中计算出的特征值,获得与输入的特征值对应的像素的相的标签,由此对图像的金相组织的相进行分类。