Automated binary processing of thermographic sequence data
AssigneeThermal Wave Imaging, Inc.
InventorSteven Shepard
Life of this patent
Estimating the term…
Abstract
A method for processing thermographic data is disclosed including thermally disturbing a specimen of unknown quality, collecting thermal data from said specimen, converting the collected data into a measure of variance and comparing the variance from the sample of unknown quality against the variance of a sample of known quality in order to determine if the quality of the specimen of unknown quality is acceptable.
Claims19
A method of processing thermal graphic data, comprising the steps of:
- thermally disturbing a first specimen of known acceptable quality,
- collecting a first collection of thermal data from said first specimen as the temperature of the first specimen temporally deviates,
- converting said first collection of data into a first measure of variance,
- thermally disturbing a second specimen of unknown quality,
- collecting a second collection of thermal data from said second specimen as the temperature of the second specimen deviates,
- converting said second collection of data into a second measure of variance,
- comparing said first and second measures of variance to one another to determine if the second measure of variance is within a predetermined range of acceptability.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said operating step further includes the step of:
- identifying discrete defects in a sample including identifying one or more pixels with a positive 2nd derivative peak followed by a negative 2nd derivative peak.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said operating step further includes of the step of:
- identifying discrete defects in a sample that fall within a defined depth range including finding one or more pixels with a positive 2nd derivative peak followed by a negative 2nd derivative peak wherein the positive 2nd derivative peaks fall within a defined time range.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said operating step further includes the step of:
- confirming that the sample is substantially similar to a predetermined standard including finding one or more pixels with a 2nd derivative peak time, peak time amplitude, and a zero crossing time behavior that substantially matches the behavior of the predetermined sample.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and operating on said converted data using at least one binary classification operation,
- wherein said operating step further includes the step of:
- identifying portions of a sample that do not match corresponding portions of the predetermined standard including finding one or more pixels of the sample that do not have a 2nd derivative peak time, peak amplitude, width, and zero crossing time behavior that matches the behavior of the predetermined standard.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said operating step further includes the step of:
- detecting anomalous boundary conditions on a back wall of the sample including identifying one or more pixels where a post-peak 2nd derivative includes an asymptotically negative portion.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation
- wherein said operating step further includes the step of:
- detecting an inclusion for thermally conductive substrate coupled to said sample including detecting one or more pixels with only a negative peak.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said converting data step further includes the step of:
- associating a peak of said 2nd derivative function with a constant width value.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said converting data step further includes the steps of:
- associating a peak of said 2nd derivative function with a constant peak height value.
The method of binary processing of thermographic data, comprising the steps of:
- thermally disturbing a sample,
- collecting thermal data from said sample,
- converting said collected thermal data using a 2nd derivative function, and
- operating on said converted data using at least one binary classification operation,
- wherein said converting data step further includes correlating the timing of a peak of said 2nd derivative to a physical attribute of a subsurface attribute of said sample.
Dependencies are inferred from the claim text.