Context-based automated defect classification system using...

Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C700S110000, C706S900000

Reexamination Certificate

active

06456899

ABSTRACT:

FIELD OF THE INVENTION
The invention relates generally to methods for semiconductor defect detection and classification. More particularly, the invention relates to a digital computer system for carrying out defect detection and defect classification operations in one automated system. The system also utilizes multiple defect masks and multiple focus planes in the defect feature extraction process.
BACKGROUND OF THE INVENTION
The automatic detection of defects during the fabrication of semiconductor wafers is largely automated, but the classification of those defects is in many cases still performed manually by technicians. Projections by semiconductor manufacturers are that with larger wafer sizes and smaller line width technology, the number of defects to be manually classified will increase exponentially. While humans are adept at classifying suspect defects, modem automated wafer inspection systems are already detecting 50,000 to 100,000 defects a day. In an attempt to reduce the volume of defects that must be investigated manually, ways are being sought to further automate the defect identification and classification processes.
In our invention, the semiconductor defect classification is based on the features of the defects themselves and also on their location with respect to process layers of the semiconductor wafer. By combining both defect features and defect location (context) features, our invention makes possible the correct classification of many types of semiconductor defects, and it allows the manual investigations as to the source of the defects to be carried out more quickly and accurately.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an automated semiconductor wafer defect classification system that classifies defects based on a combination of their location with respect to process layers of the semiconductor wafer and the defect characteristics.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that uses region classification based on defect edge information and intensity content.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that decreases wafer inspection time, improves the reliability of the defect classification, and thereby increases the process throughput and yield.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that incorporates image analysis, feature extraction, and pattern recognition and classification methods.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that utilizes image segmentation and feature extraction in the presence of semi-regular background texture.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that utilizes new defect features for better differentiating between defect types and classes.
It is another object of the invention to provide an automated semiconductor wafer defect classification system that utilizes feature extraction methods based on fractal dimension, multiple-focus imagery, multiple colorspace, and multiple defect mask implementations.
The present invention utilizes a digital computer for automatically classifying defects in a semiconductor wafer and comprises a means for identifying the background layers in a reference semiconductor wafer image resulting in a layer identification image, a means for registering a defect semiconductor wafer image to the reference semiconductor image and subsequently detecting a defect in the defect semiconductor image, a means for classifying the detected defect using features of the defect, a database for storing precomputed information and sharing information between the background layer identifying means and the detected defect classifying means, and a means for associating the classified detected defect with the layer identification image resulting in a context-based classification of the defect.


REFERENCES:
patent: 5544256 (1996-08-01), Brecher et al.
patent: 5943437 (1999-08-01), Sumie et al.
patent: 6259960 (2001-07-01), Inokuchi
patent: 6288782 (2001-09-01), Worster et al.
patent: 6292582 (2001-09-01), Lin et al.
patent: 6314379 (2001-11-01), Hu et al.
J.R.Dralla, J.C.Hoff, “Automatic Classification of Defects in Semiconductor Devices,” SPIE Conf. on Integrated Circuit Metrology, Inspection & Process Control IV, vol. 1261, 1990.
Jatko, W. B., et al, “Nonlinear Filter Derived from Topological Image Features,” Intl. Soc. for Optical Eng. Conf., Orlando, FL, Apr. 16, 1990.
Gleason, S. S., et al, “Subpixel Measurement of Image Features Based on Paraboloid Surface Fit.,” Intl. Soc. for Optical Eng. Conf., Boston, MA, Nov. 1990.
Sherman, R. et al, “An Automatic Defect Classification System for Semiconductor Wafers,” SPIE Conf. on Machine Vision Applications in Industrial Inspection, vol. 1907, 1993.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Context-based automated defect classification system using... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Context-based automated defect classification system using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Context-based automated defect classification system using... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2848631

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.