Segmentation of three phase micrograph: an automated approach

Authors:
Siddhartha Banerjee;Subhabrata Datta;Biplab Paul;Sanjoy Kumar Saha
Affiliations:
R. K. Mission Residential College, Narendrapur, India;Hooghly Engg. & Technology College, Hooghly, West Bengal, India;Jadavpur University, Kolkata, India;Jadavpur University, Kolkata, India
Venue:
Proceedings of the CUBE International Information Technology Conference
Year:
2012

Citing 4
Cited 0

Digital image processing

Digital image processing
On Topological Watersheds

Journal of Mathematical Imaging and Vision
Image Based Analysis of Complex Microstructures of Engineering Materials

International Journal of Applied Mathematics and Computer Science - Applied Image Processing
Microscope Image Processing

Microscope Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Quantitative description of micro-structure of the materials enables the analysis of material characteristics. Material under study undergoes heat and cool treatment and micro-structures thereby get revealed. Image processing has enormous potential in the analysis of microscopic images of the processed materials. Thus, the extraction of microstructures/phases present in the material forms the fundamental step towards material description. In this work, we present an automated scheme for segmenting the phases namely Ferrite, Martensite and Bainite present in the microscopic image of HSLA steel. First of all, region contours are obtained using topological watershed. Each region is then characterized as Ferrite, Martensite and Bainite based on average intensity and white grain concentration of the region. Finally, a refinement process is carried out to merge the adjacent regions of same type. As no state of the art technique is readily available, the segmentation output has been validated by the domain expert.