Anatoliy Fesenko from Herndon, VA, age 71

Us Patents

Method Of Creating A Divergence Transform For Identifying A Feature Of Interest In Hyperspectral Data
view source
US Patent:

7907762, Mar 15, 2011
Filed:

Jan 14, 2008
Appl. No.:

12/014034
Inventors:

Thomas E. Ramsay - Leesburg VA, US
Eugene B. Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
Assignee:

Guardian Technologies International, Inc. - Herndon VA
International Classification:

G06K 9/00
G06K 9/34
G06K 9/20
G06K 9/36
US Classification:

382128, 382173, 382282, 3562371, 348 86, 348125
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

Method For Determining Whether A Feature Of Interest Or An Anomaly Is Present In An Image
view source
US Patent:

8045805, Oct 25, 2011
Filed:

Mar 14, 2006
Appl. No.:

11/374612
Inventors:

Thomas E. Ramsay - Leesburg VA, US
Eugene B. Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
Assignee:

Applied Visual Sciences, Inc. - Herndon VA
International Classification:

G06K 9/00
G06K 9/66
G06K 9/46
US Classification:

382190, 382131, 382132, 382164, 382165
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

System And Method For Identifying Objects Of Interest In Image Data
view source
US Patent:

20060269135, Nov 30, 2006
Filed:

Mar 14, 2006
Appl. No.:

11/374142
Inventors:

Thomas Ramsay - Leesburg VA, US
Eugene Ramsay - Tucson AZ, US
Gerald Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - City of Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/00
US Classification:

382181000
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

System And Method For Identifying Feature Of Interest In Hyperspectral Data
view source
US Patent:

20060269140, Nov 30, 2006
Filed:

Mar 14, 2006
Appl. No.:

11/374084
Inventors:

Thomas Ramsay - Leesburg VA, US
Eugene Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/46
US Classification:

382191000
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

Method Of Creating A Divergence Transform For A Class Of Objects
view source
US Patent:

20060269161, Nov 30, 2006
Filed:

Mar 14, 2006
Appl. No.:

11/373908
Inventors:

Thomas Ramsay - Leesburg VA, US
Eugene Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - City of Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/36
US Classification:

382276000
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

Method For Characterizing An Image Source Utilizing Predetermined Color Spaces
view source
US Patent:

20080159605, Jul 3, 2008
Filed:

Mar 14, 2006
Appl. No.:

11/374189
Inventors:

Thomas E. Ramsay - Leesburg VA, US
Eugene B. Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - City of Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/00
US Classification:

382128, 382162
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

System And Method For Determining Whether There Is An Anomaly In Data
view source
US Patent:

20090010521, Jan 8, 2009
Filed:

Jan 14, 2008
Appl. No.:

12/014043
Inventors:

Thomas E. Ramsay - Leesburg VA, US
Eugene B. Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/00
US Classification:

382141
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.

System And Method For Identifying Signatures For Features Of Interest Using Predetermined Color Spaces
view source
US Patent:

20090324067, Dec 31, 2009
Filed:

Mar 14, 2006
Appl. No.:

11/374578
Inventors:

Thomas E. Ramsay - Leesburg VA, US
Eugene B. Ramsay - Tucson AZ, US
Gerard Felteau - Reston VA, US
Victor Hamilton - Ashburn VA, US
Martin Richard - Boucherville, CA
Anatoliy Fesenko - Herndon VA, US
Oleksandr Andrushchenko - Herndon VA, US
International Classification:

G06K 9/00
G06K 9/46
US Classification:

382165, 382190
Abstract:

A system and method for identifying objects of interest in image data is provided. The present invention utilizes principles of Iterative Transformational Divergence in which objects in images, when subjected to special transformations, will exhibit radically different responses based on the physical, chemical, or numerical properties of the object or its representation (such as images), combined with machine learning capabilities. Using the system and methods of the present invention, certain objects that appear indistinguishable from other objects to the eye or computer recognition systems, or are otherwise almost identical, generate radically different and statistically significant differences in the image describers (metrics) that can be easily measured.