Quantification of Alpha Lath in Ti-6Al-4V using OpenCV
Venkata Satya Surya Amaranth Karra, Amit K. Verma, Ali Guzel, Andrew Huck, Anthony D. Rollett
Abstract
Microstructure quantification is becoming an important ingredient for predicting material behavior. Here, we present a methodology to quantify 2-phase (α + β) basketweave Ti-6Al-4 V microstructures printed by a wire feed directed Energy Deposition (DED) process. The method focuses on automated quantification of features, such as α lamella thickness and volume fractions of both (α + β) phases, to ensure repeatability and to enable comparison across a wide array of images for subsequent analysis using pre-defined open access image processing libraries in the Python Language. A stereological correction was made for α-lath spacing based on the work of Collins et al. [1], while also assuming area fraction (in 2D images) as equivalent to volume fraction of (α + β) phases. Further, the results were compared with a commercial software package that uses a similar methodology. The methodology is expected to be generally applicable to lamellar microstructures and to other microstructure types via adaptation of the methodology for the features in question.