A Novel Coarse–Fine Method for Ball Grid Array Component Positioning and Defect Inspection

Abstract

With high production efficiency and robustness, automatic component pick-and-place technology is widely used in modern electronic industries. As one of the core components of this technology, the visual measurement system requires a suitable positioning and fault detection algorithm with high speed, accuracy, robustness, and generalization abilities, especially for positioning and inspecting ball grid array (BGA) components. This paper examines the online positioning and defect inspection problem of BGA components for component placement machines. Incorporating coarse and fine positioning, an accurate, efficient, and robust universal positioning algorithm is proposed. Two types of key points are introduced to characterize the alignment of solder balls in BGA components. A distance transform-based circle detection method is first applied, and then a distance-based edge filter and a circle fitting method are employed to obtain the coarse and fine locations of solder balls, respectively. The approximate location of the component is estimated using a geometrical method and the fine location is calculated by solving a least-squares problem. A pair of overlapping ratios is introduced to conduct fault detection and inspect the alignment accuracy. The effectiveness of the proposed method is verified by applying it to several real component positioning and defect inspection experiments.