Abstract
This paper presents a machine vision based electrode tip displacement measurement. The displacement is obtained by the observation of a certain area (mask) in a sequence of images. Based on the obtained displacement measurement a tip velocity was calculated as well. For this signal Kalman filtering was used in order to reduce noise. The relation between the electrode tip displacement/velocity curve parameters and weld strength was investigated. Among maximum displacement, indentation depth, maximal velocity in cooling phase and maximal velocity in heating phase the latter turned out to be the best parameter for weld strength estimation.