DocumentCode :
3016440
Title :
Initial results on the development of a new wafer inspection paradigm
Author :
El Doker, Tarek A. ; King, Josh ; Scott, David R.
Author_Institution :
Machine Perception Lab., Northern Arizona Univ., Flagstaff, AZ, USA
fYear :
2004
fDate :
28-30 March 2004
Firstpage :
124
Lastpage :
127
Abstract :
A new paradigm for wafer inspection is being developed that would resolve many of today\´s pending wafer inspection issues. This paradigm integrates 1) a DRAM fabrication line simulation model, producing synthetic images of "typical" wafer maps and associated defects, to 2) fuzzy clustering/declustering algorithms that identify various defects and 3) a unique defect tracking mechanism to monitor patterns of defects across wafer maps. This approach holds promise for in-line process control by allowing for off-site analysis of fabrication line problems and unsupervised adaptation and optimization of application-specific inspection algorithms. The paper reports on the progress made towards the fulfilment of this paradigm.
Keywords :
automatic optical inspection; fuzzy logic; image recognition; image texture; optimisation; pattern clustering; semiconductor device manufacture; DRAM fabrication line; defect identification; defect tracking mechanism; fabrication line problems; fuzzy clustering; fuzzy declustering; fuzzy logic algorithms; in-line process control; optimization; texture analysis; unsupervised adaptation; wafer inspection paradigm; wafer maps; Data mining; Fuzzy logic; Graphical user interfaces; Inspection; Laboratories; Manufacturing industries; Nearest neighbor searches; Pulp manufacturing; Random access memory; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Analysis and Interpretation, 2004. 6th IEEE Southwest Symposium on
Print_ISBN :
0-7803-8387-7
Type :
conf
DOI :
10.1109/IAI.2004.1300958
Filename :
1300958
Link To Document :
بازگشت