The need for accurate and rapid field identification of wood to combat illegal logging around the world is outpacing the ability to train personnel to perform this task. Despite increased interest in non-anatomical (DNA, spectroscopic, chemical) methods for wood identification, anatomical characteristics are the least labile data that can be extracted from solid wood products, independent of wood processing (sawing, drying, microbial attack). Wood identification using anatomical characteristics is thus still a viable approach to the wood identification problem, and automating the process of identification is an attractive and plausible solution. The undisputed increase of computer power and image acquisition capabilities, along with the decrease of associated costs, suggests that it is time to move toward non-human based automated wood identification systems and methods. This article briefly reviews the foundations of image acquisition and processing in machine vision systems and overviews how machine vision can be applied to wood identification.