Application of image analysis for characterization of spatial arrangements of features in microstructure
- PDF / 1,658,911 Bytes
- 8 Pages / 613 x 788.28 pts Page_size
- 62 Downloads / 164 Views
I.
INTRODUCTION
S P A T I A L arrangements of features in a microstructure can be quantified in terms of the distances between the corresponding features. Distribution of distances between features in a microstructure is an important microstructural attribute. The microstructural distance distributions affect the microstructural processes that involve short range, and/or long-range particle-particle interactions (or feature to feature interactions, in general). For example, diverse phenomena, such as formation and growth of microcracks during thermal cycling of metal matrix composites, t~,21 damage accumulation and fracture behavior of composites, t3,41 particle coarsening, rS,< liquid-phase sintering, t8,9~ microstructural evolution during solid-state transformations where nucleation occurs on the second-phase particles, tl~ microvoid induced ductile fracture processes in metals and alloys, [lz,~3] creep cavitation, t~4Jand creep crack growth, t~51 crucially depend on the distances between the relevant microstructural features. The microstructural distance distributions may be significantly altered when a material is processed in reduced gravity or microgravity, particularly if a liquid phase is involved. In such a case, it is desirable to quantify the changes in the microstructural distance distributions due to reduced gravity and to quantify and model the role of gravity in the microstructural evolution processes. Very few investigations deal with the quantitative effect of the spatial distribution of microstructural features on the physical/mechanical properties of materials or on the microstructural evolution during materials processing. An important reason for this is the lack of welldeveloped practical techniques to quantify the spatial distribution of microstructural features and, as a result, the lack of realistic and flexible quantitative models to represent the spatial distribution in the computer simulation or analytical theoretical studies. PASCAL LOUIS, Research Engineer, is with Aerospatiale, Centre Commun de Recherches, 92152 SURESNES Cedex, France. ARUN M. GOKHALE, Professor, is with the School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA 303320245. Manuscript submitted September 2, 1994. METALLURGICAL AND MATERIALS TRANSACTIONS A
It is the purpose of this article to present an image analysis procedure to obtain the basic experimental data necessary to quantify the spatial arrangement of microstructural features in metallographic sections. To characterize spatial distributions, the necessary basic data consist of centroid coordinates and sizes of the microstructural features of interest. These raw data can be processed to compute the descriptors of spatial order, such as nearest-neighbor distribution, IJ6,~7,~81 radial distribution, 119-221 K function, L19-221 pair-correlation function, t21,22,23J or parameters such as short range and long range microstructural gradients, r24] An image analysis procedure was developed to automatically extract centroid coordi
Data Loading...
