A new method for locating solder joint based on Rough Set

Continual increases in IC-chip complexity and performance are placing demands on the density and functionality of package I/Os. Therefore, various SMT (Surface Mounting Technology) interconnection techniques are being developed to satisfy this need. Non-destructive inspection technology based on image captured by CCD (Charge Couple Device) has been widely used in a production of PCB (Printed Circuit Board), because of their excellent characters. This paper deals with the locating of solder joints in PCB. In recent years, the theory of Rough Sets has gained paramount applicability in diverse areas of research, especially in Data Mining, Knowledge Discovery, Artificial intelligence and Information Systems analysis. Rough Sets have also been used in image processing; however, the application of Rough Sets for image locating analysis has yet to be fully investigated. In this paper image segmentation using Rough Set Theory has been presented, and then locate the solder joint by calculating the correlation of solder joint image. In traditional production, the locating in automatic optical inspection must be done by manual, so design a new method to find the solder joint region becomes very important. This paper discusses the flow of locating solder joint based on the rough set. Experimental results reveal that the proposed method shows practical value in solder joint inspection.