Image analysis – Applications – Manufacturing or product inspection
Reexamination Certificate
2000-03-06
2002-11-19
Patel, Jayanti K. (Department: 2721)
Image analysis
Applications
Manufacturing or product inspection
C382S218000, C382S224000
Reexamination Certificate
active
06483938
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
This invention relates to defect classification and diagnosis of manufacturing defects.
BACKGROUND OF THE INVENTION
In most manufacturing processes, management of through-put and yield are of concern. The ability to locate potential problems, identify problems, and take corrective action to obviate the source of the defect, and if possible, to repair the defect, can make a significant difference in the performance of manufacturing process. Therefore, it is desirable to have the best systems possible for identifying possible problems or anomalies, identifying an anomaly as a particular type of defect, identifying the source of the defect, and repairing the manufactured object to correct the defect if possible. This is particularly true in the semiconductor industry.
In the semiconductor manufacturing industry, a challenge remains to improve yields as the designs get smaller and smaller. Particles and process defects can limit yields in manufacturing semiconductor devices. Therefore, systems that perform the general functions described above can become extremely important. Conventional techniques have shortcomings including less than desirable speed and accuracy. With respect to identifying defects in the manufacturing process, manual classification has been required of anomalies and manual diagnosing of the cause of defects. Such manual inputs may have resulted in inconsistent results and consumption of considerable operator time.
SUMMARY OF THE INVENTION
According to an aspect of the present invention, a method for generating a knowledgebase for use in labeling anomalies on a manufactured object includes capturing an image of the object having an anomaly; preparing a pixel-based representation of the image; decomposing the pixel-based representation of the image into a primitives-based representation of the image; isolating the anomaly on the primitives-based representation of the image; comparing the primitive-based representation of the image with primitive sets of known anomalies in a knowledge base to locate the primitive set having a maximum similarity; presenting to an operator a label associated with the set of primitives having a maximum similarity to an operator; entering a label to be associated with the primitive-based representation of the image.
According to another aspect of the present invention a method for indexing information about defects includes using operating system subdirectories names as defect attributes and producing compact indexes of the contents of defect files by use of operating-system commands to produce an index of the subdirectory names in an object-oriented format in order to provide fast and flexible retrieval of defect information without having to generate database tables and queries.
According to another aspect of the present invention, a method for augmenting a knowledgebase for use in labeling anomalies on a manufactured object includes capturing an image of the object having an anomaly; preparing a pixel-based representation of the image; decomposing the pixel-based representation of the image into a primitives-based representation of the image; isolating the anomaly on the primitives-based representation of the image; comparing the primitives-based representation with primitive sets in a knowledgebase to find the primitive set with a maximum similarity; obtaining a first label associated with the primitive set having a maximum similarity; associating the first label with the primitives-based representation of the image if the similarity is greater than a predetermined similarity threshold; and adding the primitive-based representation and associated first label to the knowledgebase.
According to another aspect of the present invention, a system for generating a knowledgebase for use in labeling anomalies on a manufactured object includes an image-capturing device for capturing an image of the object having an anomaly; a pixel-generating device for preparing a pixel-based representation of the image; and a computer having a processor and memory coupled to the means for preparing a pixel-based representation, the computer programmed to be operable to: decompose the pixel-based representation of the image into a primitives-based representation of the image, isolate the anomaly on the primitives-based representation of the image; store the primitive-based representation of the image, and associate an assigned label with the stored primitive based representation of the image.
According to another aspect of the present invention, rules to a knowledgebase are changed based on their ability to achieve acceptable results. According to another aspect of the present invention, a method of accumulation and assimilation of rules into a knowledgebase includes adding new rules, eliminating duplicate rules, deleting improper rules, dynamically assigning weights to descriptors based on their role in achieving acceptable results and deleting rules that do not produce acceptable results at any time with recompilation of the knowledgebase. According to another aspect of the present invention, a knowledgebase is enhanced to promote efficiency.
REFERENCES:
patent: 4233625 (1980-11-01), Atman
patent: 4353087 (1982-10-01), Berry et al.
patent: 4509075 (1985-04-01), Simms et al.
patent: 4550374 (1985-10-01), Meshman et al.
patent: 4579455 (1986-04-01), Levy et al.
patent: 4641257 (1987-02-01), Ayata
patent: 4791586 (1988-12-01), Maeda et al.
patent: 4794648 (1988-12-01), Ayata et al.
patent: 4907285 (1990-03-01), Nakano et al.
patent: 4955062 (1990-09-01), Terui
patent: 5048100 (1991-09-01), Kuperstein
patent: 5109430 (1992-04-01), Nishihara et al.
patent: 5144683 (1992-09-01), Suzuki et al.
patent: 5220621 (1993-06-01), Saitoh
patent: 5311600 (1994-05-01), Aghajan et al.
patent: 5321770 (1994-06-01), Huttenlocher et al.
patent: 5436979 (1995-07-01), Gray et al.
patent: 5515453 (1996-05-01), Hennessey et al.
patent: 5544256 (1996-08-01), Brecher et al.
patent: 5563702 (1996-10-01), Emery et al.
patent: 5659630 (1997-08-01), Forslund
patent: 5701394 (1997-12-01), Arita et al.
patent: 5748872 (1998-05-01), Norman
patent: 5801965 (1998-09-01), Takagi et al.
patent: 5808735 (1998-09-01), Lee et al.
patent: 5875108 (1999-02-01), Hoffberg et al.
“Investigation of a Fuzzy Grammar for Automated Visual Inspection,” by Hahn Kwang-Soo, Dissertation in Interdisciplinary Engineering, Graduate Faculty, Texas Tech University, Aug. 1989.
“Picture Interpretation A Symbolic Approach,” by Sandy Dance, Terry Caelli and Zhi-Qiang Liu, Series in Machine Perception and Artificial Intelligence—vol. 20, World Scientific, Apr. 1995.
“Symbolic Image Interpretation by Parsing, Interpreting, and Pruning,” by Staffan Truvé, A Dissertation for the Ph.D. Degree in Computing Science at Chalmers University of Technology, Department of Computer Sciences, dated Jun. 4, 1992.
“Symbolic Reasoning Among 3-D Models and 2-D Images,” A Dissertation Submitted to the Department of Computer Science and the Committee on Graduate Studies of Stanford University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy, by Rodney Allen Brooks, dated Jun. 1981.
“Symbolic Pixel Labeling for Curvilinear Feature Detection,” Computer Vision Laboratory, Center for Automation Research, University of Maryland, by John Canning, J. John Kim, and Azriel Rosenfeld, dated Jan. 1987.
“Symbolic Learning and Interpretation of 2-D Electrophoresis Gels,” University Libraries, Carnegie Mellon University, by Pierre Nugues, Robert Whalen and Jean-Paul Haton, dated Apr. 11, 1991.
“Techniques for Syntactic Analysis of Images with Application for Automatic Visual Inspection,” Dissertation in Interdisciplinary Engineering, Graduate Faculty, Texas Tech University, by Youling Lin, Dec. 1990.
“Segmentation Through Symbolic Surface Descriptions,” Computer Vision Laboratory, Electrical Engineering and Computer Science Dept., The University of Michigan, by Paul Besl and Ramesh Jain, dated Feb. 1986.
“Zero-Crossing Symbolic Vision Primitives Emulating Physiologic Encoding Scheme
Cleavelin C. Rinn
Hastings, II Howard V.
Hennessey A. Kathleen
Kinikoglu Pinar
Lin YouLing
Patel Jayanti K.
Telecky , Jr. Frederick J.
Texas Instruments Incorporated
Troike Robert L.
LandOfFree
System and method for classifying an anomaly does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for classifying an anomaly, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for classifying an anomaly will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2964680