Processing and Refinement of Steel Microstructure Images for Assisting in Computerized Treatment of Plain Carbon Steel

Abstract

In this thesis, we propose a technique to refine raw microstructure images of plain carbon steel, into a digital form usable in experiments related to heat treatment processes of plain carbon steel for diverse applications. The steel microstructure refinement technique proposed in this thesis is aimed towards the development of a database of digital steel images, which will be beneficial for materials engineers to perform computer–aided simulations of heat treatment of plain carbon steel. Simulations using these refined images would enable automation of steel treatment processes and hence would save time and resource requirements considerably, as compared to extensive physical experimentations, while producing almost equally accurate results. The proposed method involves image filtering, segmentation, achieving edge uniformity, and essentially consists of six broad steps. At each step, we considerably expunge one or more of undesired properties of the raw microstructure image. Our experimental results prove the efficiency and effectiveness of the proposed technique.