Regresar

Metallographic Analysis for Low Carbon Steels Using Software Developed with Python

Abstract:

This research focuses on automating the determination of grain size, the percentage of ferritic-pearlitic phases in low carbon steels by developing a metallographic analysis software programmed in Python. The direct count valuation method and the values of the G granulometric indices according to the UNE 7-280.72 standard are used. The process consisted of taking the micrographs of 4 types of steels with different carbon content, each type of steel is made with 10 micrographs with a 100x zoom, once the images are entered into the software, the program segments the light and dark regions of the image and counts the total regions and the dark regions (ferrite and pearlite) and compares them with the standard UNE. The results of the investigation were compared with the results of the commercial metallographic software PAX-it and using an ANOVA analysis showed that the differences between the results for the by grain size and the percentage of ferrite obtained with the developed software and the PAX-it they are not statistically significant and present a confidence level of 95%, validating the results.