Hough transform for the calculation of twist angle of aramid torsion

Abstract

Aramid yarns are widely used for various technical applications. For example, they are used for creation of a magnetosensitive element of the torsion magnetometer, which is a part of the geophysical complex GI-MTS-1 (SPbF IZMIRAN). Such element is made of three microfilaments of aramid yarn. Study of such physicalmechanical characteristics of yarns as tensile, shear and torsion moduli, is necessary for improvement torsion magnetometer. This paper describes the preparatory phase of experimental study of yarn twist angle variation during tensile test. It includes equipment description, in particular, description of the scheme of video capture, and developed computer vision algorithm for determining an aramid torsion slope angles. Proposed scheme of video capture includes high-speed camera and allows precise focusing due to two-coordinate platform and avoiding of optical distortions. A special computer vision algorithm, which based on the Hough transform, was developed. This algorithm was tested on torsion images obtained with digital microscope. The algorithm works with different real yarns regardless of their color and twist type.