Image analysis to measure strain in wood and paper

Abstract

A fast and accurate strain measurement method was developed to quantify microscopic deformation in wood and paper using video images. Two or more video images, one of which was undeformed and the others of which were deformed, were digitized and compared using the image correlation technique. This digital computer pattern recognition technique permitted the displacement of any point in the image to be measured. Then, strain between any two points in this observation field could be calculated and, hence, normal, shearing strains and Poisson's ratio could be measured directly. Independent strain measurements showed that the method using video images provided accurate strain measurements and could be applied to complex materials like wood and wood products. The strain measurements using the image analysis produced not only normal strains but strain profiles of the normal strains. From the strain profiles, it was possible to detect local areas of strain concentration which might be beyond the elastic range. As it measures displacements and strains from a set of video images using the image correlation technique, the method is non-contact and the observation field can be adjusted easily by changing magnifying power of the optical devices. Thus, the method can be applied to mechanical testing of wood and wood products from the cellular level to full size lumber and will no longer be limited by the fragile nature of the material being tested, industry standards, or equipment restrictions.