Semiconductor yield improvements through automatic defect classification - Final Report
Automatic detection of defects during the fabrication of semiconductor wafers is largely automated, but the classification of those defects is still performed manually by technicians. Projections by semiconductor manufacturers predict that with larger wafer sizes and smaller line width technology the number of defects to be manually classified will increase exponentially. This cooperative research and development agreement (CRADA) between Martin Marietta Energy Systems (MMES) and KLA Instruments developed concepts, algorithms and systems to automate the classification of wafer defects to decrease inspection time, improve the reliability of defect classification, and hence increase process throughput and yield. Image analysis, feature extraction, pattern recognition and classification schemes were developed that are now being used as research tools for future products and are being integrated into the KLA line of wafer inspection hardware. An automatic defect classification software research tool was developed and delivered to the CRADA partner to facilitate continuation of this research beyond the end of the partnership.
- Research Organization:
- Oak Ridge Y-12 Plant (Y-12), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC05-84OS21400
- OSTI ID:
- 508162
- Report Number(s):
- Y/AMT-423; CRADA ORNL92-0140; ON: DE97008098; CRN: C/ORNL--92-0140; TRN: 97:004673
- Resource Relation:
- Other Information: PBD: 30 Sep 1995
- Country of Publication:
- United States
- Language:
- English
Similar Records
Context-based automated defect classification system using multiple morphological masks
Automatic classification of spatial signatures on semiconductor wafermaps