Patent · US Active

Part replacement predictions using convolutional neural networks

US11523004B2 · kind B2 · utility

0Cited by

12References

15Claims

0Family size

Assignees

Inventors

Darrel Cherry · Meridian, US
Renee J. Jessome · Boise, US
Richard E. Maggard · Boise, US
Daniel Siddall · Boise, US
Yeshwanth Raja · Rochester, US
Eli Saber · Webster, US

Key dates

Filing date	Sep 21, 2018
Grant date	Dec 6, 2022
Priority date	—
Expiry date	Sep 21, 2038

Classification

Technology area (CPC G)Physics
CPC primaryG06F11/3409
WIPO fieldAudio-visual technology
WIPO sectorElectrical engineering

Abstract

An example of an apparatus including a communication interface to receive an image file is provided. The image file represents a scanned image of a output generated by a printing device. The apparatus further includes an identification engine to process the image file with a convolutional neural network model to identify a feature. The feature may be indicative of a potential failure. The apparatus also includes an image analysis engine to indicate a life expectancy of a part associated with the potential failure based on the feature. The image analysis engine uses the convolutional neural network model to determine life expectancy.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.