In automatic defect classification of semiconductor wafers using scanning electron microscope images, we propose a technique of tuning decision parameters for rule‐based defect classifiers. The proposed method adopts a coarse‐to‐fine search for reduction in processing time. However, due to search leakage, there is no guarantee that the same solutions as in a full parameter search can be obtained. In order to prevent leakage in the coarse search theoretically, the proposed method evaluates a candidate parameter set based on the estimated range of classification accuracy attained by not only the candidate but the surrounding solutions eliminated by the coarse search. Experiments on real image data demonstrate the effectiveness of the proposed method. The proposed method can extract the same solutions as the full parameter search within almost the same processing time as the conventional coarse‐to‐fine search. When the sampling step of the coarse search is three to six, while the tuning time of the conventional coarse‐to‐fine search is 1 to 21 s, that of the proposed method is 5 to 35 s.
A Technique of Automatically Tuning Decision Parameters for Rule‐Based Defect Classifier
Electrical engineering in Japan ; 196 , 1
2016
Aufsatz (Zeitschrift)
Englisch
A HEDGE ALGEBRAS BASED REASONING METHOD FOR FUZZY RULE BASED CLASSIFIER
BASE | 2019
|Decision Support for Rule and Technique Discovery in an Uncertain Environment
British Library Conference Proceedings | 2002
|A Stable Method for Detecting Driver Maneuvers Using a Rule Classifier
Springer Verlag | 2022
|