Vision Character And Quality Information Identify Of MAG Welding

The purpose of the subject was to acquire the useful quality characteristic informationfrom the weld pool image, which built a good foundation for the welding quality controlbased on vision sensing.A passive vision sensing system for taking the image of MAG welding pool had beenset up in this paper. CCD and the near-infrared compound filters system were used toeliminate the arc disturbance and a lot of clear images were obtained. Image-processingsoftware was used to analyse the gray character. Gray mean was put forward to study theprocess of MAG welding. Taking the weld pool outline as object, the weld pool geometryparameter system was established and the relationship between the weld pool edgecharacter and the welding stability was analyzed. Taking the interior pool as the object, thearc and the molten-slag's contours were extracted.The relationship between image characteristic information and the weld defect wasstudied by sync contrast experiments. The visual characters of the pore, weld collapse andthe other defect were extracted. The experiments approved that one kind of image maypredicate several kinds of welding defect, one kind of welding defect may also showedseveral kinds of image characters. Experimental results show that it is feasible to judge thedefect by visual characters which offered a technological foundation for automaticidentification of welding defect on the base of vision sensing.