Automatic morphological extraction of fibers from SEM images for quality control of short fiber-reinforced composites manufacturing

Abstract

The properties of fiber-reinforced composite materials greatly depend on the morphology of reinforcing fibers within the base materials, i.e., spatial uniformity, orientation, and length distribution. Accurately extracting this information in an automated manner from SEM images is essential for quality assessment, quality control, and process optimization. However, due to overlapping or cross-linking issue, morphological fiber extraction is very challenging and has not been well addressed in the existing literature. This paper takes into account this research gap and proposes five different methods, namely, the opening method, simple Hough transform (HT), partitioning HT, gradient-based HT, and break-merge method to automatically extract the straight fibers from SEM images to facilitate the morphological analysis. The robustness and accuracy of the proposed methods are evaluated and compared through both simulation and real case studies.