30 December 2016 Classification automation of thermoplastic particles in a cured epoxy matrix according to their size on microscopic images
Author Affiliations +
Proceedings Volume 10224, International Conference on Micro- and Nano-Electronics 2016; 102242B (2016) https://doi.org/10.1117/12.2267090
Event: The International Conference on Micro- and Nano-Electronics 2016, 2016, Zvenigorod, Russian Federation
Abstract
Epoxy resins have wide applications in modern industries. To improve the properties of such resins the thermoplastic component is often used. This component dissolves in the epoxy resin at a high temperature. To determine the properties of the obtained cured epoxy matrix with thermoplastic particles it is important to estimate and classify this particle sizes. In this paper we investigate methods for solving these tasks automatically. The thermoplastic particle sizes are analyzed using the microscopy images of the cured epoxy matrix. The digital image processing methods for the thermoplastic particle detection are discussed. The Otsu’s method is implemented for microscopic images with homogeneous background. The Circular Hough Transform method is implemented for microscopic images with big visible particle radii. The results of both methods for the considered images are represented. The parameters of the Gaussian distribution for the thermoplastic particle sizes in a cured epoxy matrix are estimated from the analyzed microscopic images.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Victoria A. Sablina, Alexander N. Varnavsky, Andrei N. Varnavsky, "Classification automation of thermoplastic particles in a cured epoxy matrix according to their size on microscopic images", Proc. SPIE 10224, International Conference on Micro- and Nano-Electronics 2016, 102242B (30 December 2016); doi: 10.1117/12.2267090; https://doi.org/10.1117/12.2267090
PROCEEDINGS
6 PAGES


SHARE
Back to Top