Method for localizing and isolating an errant process step
Abstract
A method for localizing and isolating an errant process includes the steps of retrieving from a defect image database a selection of images each image having image content similar to image content extracted from a query image depicting a defect, each image in the selection having corresponding defect characterization data. A conditional probability distribution of the defect having occurred in a particular process step is derived from the defect characterization data. A process step as a highest probable source of the defect according to the derived conditional probability distribution is then identified. A method for process step defect identification includes the steps of characterizing anomalies in a product, the anomalies detected by an imaging system. A query image of a product defect is then acquired. A particular characterized anomaly is then correlated with the query image. An errant process step is then associated with the correlated image.
- Inventors:
-
- (Harriman, TN)
- Knoxville, TN
- Issue Date:
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- OSTI Identifier:
- 875118
- Patent Number(s):
- 6535776
- Assignee:
- Ut-Battelle, LLC (Oak Ridge, TN)
- Patent Classifications (CPCs):
-
G - PHYSICS G01 - MEASURING G01N - INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
G - PHYSICS G01 - MEASURING G01R - MEASURING ELECTRIC VARIABLES
- DOE Contract Number:
- AC05-96OR22464
- Resource Type:
- Patent
- Country of Publication:
- United States
- Language:
- English
- Subject:
- method; localizing; isolating; errant; process; step; steps; retrieving; defect; image; database; selection; images; content; similar; extracted; query; depicting; characterization; data; conditional; probability; distribution; occurred; derived; probable; source; identified; identification; characterizing; anomalies; product; detected; imaging; acquired; characterized; anomaly; correlated; associated; image data; /700/382/
Citation Formats
Tobin, Jr., Kenneth W., Karnowski, Thomas P, and Ferrell, Regina K. Method for localizing and isolating an errant process step. United States: N. p., 2003.
Web.
Tobin, Jr., Kenneth W., Karnowski, Thomas P, & Ferrell, Regina K. Method for localizing and isolating an errant process step. United States.
Tobin, Jr., Kenneth W., Karnowski, Thomas P, and Ferrell, Regina K. Wed .
"Method for localizing and isolating an errant process step". United States. https://www.osti.gov/servlets/purl/875118.
@article{osti_875118,
title = {Method for localizing and isolating an errant process step},
author = {Tobin, Jr., Kenneth W. and Karnowski, Thomas P and Ferrell, Regina K},
abstractNote = {A method for localizing and isolating an errant process includes the steps of retrieving from a defect image database a selection of images each image having image content similar to image content extracted from a query image depicting a defect, each image in the selection having corresponding defect characterization data. A conditional probability distribution of the defect having occurred in a particular process step is derived from the defect characterization data. A process step as a highest probable source of the defect according to the derived conditional probability distribution is then identified. A method for process step defect identification includes the steps of characterizing anomalies in a product, the anomalies detected by an imaging system. A query image of a product defect is then acquired. A particular characterized anomaly is then correlated with the query image. An errant process step is then associated with the correlated image.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed Jan 01 00:00:00 EST 2003},
month = {Wed Jan 01 00:00:00 EST 2003}
}
Works referenced in this record:
Content based image retrieval systems
journal, January 1995
- Gudivada, V. N.; Raghavan, V. V.
- Computer, Vol. 28, Issue 9
A knowledge-based approach for retrieving images by content
journal, August 1996
- Hsu, Chic-Cheng; Chu, W. W.; Taira, R. K.
- IEEE Transactions on Knowledge and Data Engineering, Vol. 8, Issue 4, p. 522-532
Indexing pictorial documents by their content: a survey of current techniques
journal, February 1997
- De Marsicoi, M.; Cinque, L.; Levialdi, S.
- Image and Vision Computing, Vol. 15, Issue 2