Active Shape Model for Visual Detection and Localization of Variable Polarity PAW Weld Pool

In order to control the quality process of variable polarity PAW weld of aluminum alloys, extracting its characteristic parameters of weld pool is an important issue for automated and robotic welding, wherein the visual detection and localization of weld pool are critical steps. Previous methods of visual measurement of weld pool usually depend on artificial experience to detect weld pool and assume that the weld pool exists in a special shape and size. Considering these methods are based on idealized model and lack of enough accuracy, in this paper, we propose a robust visual detection and localization method for the pool of variable polarity PAW weld pool based on active shape model. It provides an elegant framework for representing the outline of a weld pool and is especially efficient for weld pool detection and localization in cluttered scenes. The result shows, it can detect and localize weld pool faster and more efficiently which lay a solid foundation for extraction of characteristic parameters.