Using fuzzy mathematics to detect dimple defects of polished wafer surface

Abstract

Using the concept of fuzzy mathematics, the authors developed an automated visual inspection system to detect dimple defects of polished wafer surfaces. The algorithm consists of two major processing phases. At the first phase, pre-processing is performed to eliminate noise and to reduce the number of potential candidates of dimple defects. At the second phase, four pattern features are defined based on the consideration of scale-, position-, and orientation-invariant. A fuzzy membership function is utilized to cope with the wide range of shape variations of the dimple defects. A decision-making mechanism is based on the value of the membership function which describes the pattern's closeness to a dimple. The attractive features of the system include the fact that the algorithm is distortion-invariant. Experimental results are presented.<>