Abstract

Titanium is a material that exhibits many desirable properties including a very high strength to weight ratio and corrosive resistance. However, the specific properties of any components depend upon the microstructure of the material, which varies by the manufacturing process. This means it is often necessary to analyse the microstructure when designing new processes or performing quality assurance on manufactured parts. For Ti6Al4V, grain size analysis is typically performed manually by expert material scientists as the complicated microstructure of the material means that, to the authors knowledge, no existing software reliably identifies the grain boundaries. This manual process is time consuming and offers low repeatability due to human error and subjectivity. In this paper, we propose a new, automated method to segment microstructural images of a Ti6Al4V alloy into its constituent grains and produce measurements. The results of applying this technique are evaluated by comparing the measurements obtained by different analysis methods. By using measurements from a complete manual segmentation as a benchmark we explore the reliability of the current manual estimations of grain size and contrast this with improvements offered by our approach.

Fingerprint

Titanium alloys
Image processing
Microstructure
Caustics
Titanium
Quality assurance
Grain boundaries
titanium alloy (TiAl6V4)

Keywords

  • titanium
  • alloys
  • digital image processing
  • microstructural images

Cite this

@article{259b5bff51354297bd8e53cace717ef9,
title = "Automated microstructural analysis of titanium alloys using digital image processing",
abstract = "Titanium is a material that exhibits many desirable properties including a very high strength to weight ratio and corrosive resistance. However, the specific properties of any components depend upon the microstructure of the material, which varies by the manufacturing process. This means it is often necessary to analyse the microstructure when designing new processes or performing quality assurance on manufactured parts. For Ti6Al4V, grain size analysis is typically performed manually by expert material scientists as the complicated microstructure of the material means that, to the authors knowledge, no existing software reliably identifies the grain boundaries. This manual process is time consuming and offers low repeatability due to human error and subjectivity. In this paper, we propose a new, automated method to segment microstructural images of a Ti6Al4V alloy into its constituent grains and produce measurements. The results of applying this technique are evaluated by comparing the measurements obtained by different analysis methods. By using measurements from a complete manual segmentation as a benchmark we explore the reliability of the current manual estimations of grain size and contrast this with improvements offered by our approach.",
keywords = "titanium, alloys, digital image processing, microstructural images",
author = "A. Campbell and P. Murray and E. Yakushina and S. Marshall and W. Ion",
year = "2017",
month = "3",
day = "1",
doi = "10.1088/1757-899X/179/1/012011",
language = "English",
volume = "179",
journal = "IOP Conference Series: Materials Science and Engineering",
issn = "1757-8981",
number = "1",

}

TY - JOUR

T1 - Automated microstructural analysis of titanium alloys using digital image processing

AU - Campbell, A.

AU - Murray, P.

AU - Yakushina, E.

AU - Marshall, S.

AU - Ion, W.

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Titanium is a material that exhibits many desirable properties including a very high strength to weight ratio and corrosive resistance. However, the specific properties of any components depend upon the microstructure of the material, which varies by the manufacturing process. This means it is often necessary to analyse the microstructure when designing new processes or performing quality assurance on manufactured parts. For Ti6Al4V, grain size analysis is typically performed manually by expert material scientists as the complicated microstructure of the material means that, to the authors knowledge, no existing software reliably identifies the grain boundaries. This manual process is time consuming and offers low repeatability due to human error and subjectivity. In this paper, we propose a new, automated method to segment microstructural images of a Ti6Al4V alloy into its constituent grains and produce measurements. The results of applying this technique are evaluated by comparing the measurements obtained by different analysis methods. By using measurements from a complete manual segmentation as a benchmark we explore the reliability of the current manual estimations of grain size and contrast this with improvements offered by our approach.

AB - Titanium is a material that exhibits many desirable properties including a very high strength to weight ratio and corrosive resistance. However, the specific properties of any components depend upon the microstructure of the material, which varies by the manufacturing process. This means it is often necessary to analyse the microstructure when designing new processes or performing quality assurance on manufactured parts. For Ti6Al4V, grain size analysis is typically performed manually by expert material scientists as the complicated microstructure of the material means that, to the authors knowledge, no existing software reliably identifies the grain boundaries. This manual process is time consuming and offers low repeatability due to human error and subjectivity. In this paper, we propose a new, automated method to segment microstructural images of a Ti6Al4V alloy into its constituent grains and produce measurements. The results of applying this technique are evaluated by comparing the measurements obtained by different analysis methods. By using measurements from a complete manual segmentation as a benchmark we explore the reliability of the current manual estimations of grain size and contrast this with improvements offered by our approach.

KW - titanium

KW - alloys

KW - digital image processing

KW - microstructural images

UR - http://www.scopus.com/inward/record.url?scp=85016494499&partnerID=8YFLogxK

U2 - 10.1088/1757-899X/179/1/012011

DO - 10.1088/1757-899X/179/1/012011

M3 - Article

VL - 179

JO - IOP Conference Series: Materials Science and Engineering

T2 - IOP Conference Series: Materials Science and Engineering

JF - IOP Conference Series: Materials Science and Engineering

SN - 1757-8981

IS - 1

M1 - 012011

ER -