Scott T Roberts

US Patent:

20100258149, Oct 14, 2010

Filed:

May 28, 2009

Appl. No.:

12/474237

Inventors:

Chris Schneller - Tijeras NM, US
Scott Roberts - Albuquerque NM, US
Kathryn Whitaker - Albuquerque NM, US
Kevin Ferguson - Albuquerque NM, US
Stephen Islas - Albuquerque NM, US
Joe Moore - Austin TX, US
Jim Puissant - Albuquerque NM, US

Assignee:

Poly-Flow Engineering, LLC - Albuquerque NM

International Classification:

B08B 9/00

US Classification:

134170, 134 221

Abstract:

Wafer carrier washing and drying apparatus and method, especially useful for the semiconducting industry.

US Patent:

61640352, Dec 26, 2000

Filed:

Nov 23, 1998

Appl. No.:

9/198123

Inventors:

Scott J. Roberts - Flagstaff AZ

International Classification:

E04B 112

US Classification:

52563

Abstract:

A foam wall assembly including vertical passageways that guide wall support elements. The wall assembly includes a lower end and an upper support element that are affixed to the wall support elements. The foam wall includes inner and outer thermal barriers that thermally isolate the wall support elements.

US Patent:

58392491, Nov 24, 1998

Filed:

Oct 16, 1996

Appl. No.:

8/730940

Inventors:

Scott J. Roberts - Flagstaff AZ

International Classification:

E04B 100

US Classification:

5274508

Abstract:

Foam blocks including vertical passageways are aligned by a block alignment element having first and second edge alignment surfaces for simultaneously engaging the internal block passageways at angularly spaced apart intervals to align one block relative to another block.

US Patent:

20220301291, Sep 22, 2022

Filed:

Jun 3, 2022

Appl. No.:

17/832477

Inventors:

- Albuquerque NM, US
Kevin Matthew Potter - Albuquerque NM, US
Emily Donahue - Albuquerque NM, US
Matthew David Smith - Albuquerque NM, US
Charles J. Snider - Albuquerque NM, US
John P. Korbin - Albuquerque NM, US
Scott Alan Roberts - Albuquerque NM, US
Lincoln Collins - Albuquerque NM, US

International Classification:

G06V 10/776
G06T 7/11
G06V 10/82
G06V 10/764

Abstract:

Digital image segmentation is provided. The method comprises training a neural network for image segmentation with a labeled training dataset from a first domain, wherein a subset of nodes in the neural net are dropped out during training. The neural network receives image data from a second, different domain. A vector of N values that sum to 1 is calculated for each image element, wherein each value represents an image segmentation class. A label is assigned to each image element according to the class with the highest value in the vector. Multiple inferences are performed with active dropout layers for each image element, and an uncertainty value is generated for each image element. Uncertainty is resolved according to expected characteristics. The label of any image element with an uncertainty above a threshold is replaced with a new label corresponding to a segmentation class based on domain knowledge.

US Patent:

20210374968, Dec 2, 2021

Filed:

May 29, 2020

Appl. No.:

16/887311

Inventors:

- Albuquerque NM, US
Kevin Matthew Potter - Albuquerque NM, US
Emily Donahue - Albuquerque NM, US
Matthew David Smith - Albuquerque NM, US
Charles J. Snider - Albuquerque NM, US
John P. Korbin - Albuquerque NM, US
Scott Alan Roberts - Albuquerque NM, US
Lincoln Collins - Albuquerque NM, US

International Classification:

G06T 7/174
G06N 3/08
G06N 5/04
G06K 9/62
G06F 17/18

Abstract:

A method for digital image segmentation is provided. The method comprises training a neural network for image segmentation with a labeled training dataset from a first domain, wherein a subset of nodes in the neural net are dropped out during training. The neural network receives image data from a second, different domain. A vector of N values that sum to 1 is calculated for each image element, wherein each value represents an image segmentation class. A label is assigned to each image element according to the class with the highest value in the vector. Multiple inferences are performed with active dropout layers for each image element, and an uncertainty value is generated for each image element. The label of any image element with an uncertainty value above a predefined threshold is replaced with a new label corresponding to the class with the next highest value.