Machine vision recognition of disconnection failure of IC wafer

The defect exhibited a great variety of kinds have a direct impact in the yield of IC wafer in the lithography process. The missing material defect is an important factor of reducing the yield of IC, it mainly causes circuits to be disconnection failure, so it is important to recognize the missing material defect and ascertain the circuit's failure style. In this paper, a recognition method for disconnection failure of the missing material defect based on skeleton character is presented. The thinning algorithm is used to extract the skeleton of image. The principle of deleting redundant points and branches is introduced to get the skeleton of single pixel and single direction. Then, the method of neighborhood field tracking is used to recognizing the disconnection circuits.