Integration of industrial photogrammetry and neuro-fuzzy system for measuring and modeling deformation of industrial product surfaces under external forces

Abstract

Deformation of surfaces under external loadings greatly is function of physical laws. But according to the impossibility of identifying all effective factors and modeling their interactions parametrically, analytical methods do not provide good performance generally. So, it is necessary to apply practical methods by carrying out field tests and measuring deformations directly. To achieve this purpose it is required to integrate capabilities of an accurate measurement technique and a flexible modeling method. The capabilities such as: high accuracy and speed in measuring 3D coordinates of desired points, ability to perform measurement in a continuous space and no need to contact with the surface of objects at the time of measurement make close range photogrammetry a reliable tool for measuring geometric parameters of an object before and after deformation. The ability to measure geometric parameters of an object before and after deformation in one hand and following the deformation from physical laws on the other hand make neuro-fuzzy system the first choice for modeling the deformation of objects using outputs of close range photogrammetry.In this paper, a new method has been presented for measuring and modeling deformation of industrial product surfaces under external forces using close range photogrammetry (as an image-based measurement tool) and neuro-fuzzy network (as a behavior modeling tool).