IP Library › Granted Patent US 11,222,402
Granted Patent B2
US 11,222,402 · App. 16/858,234 · Granted Jan 11, 2022

Adaptive image enhancement

Inventors: Narges Afsham (Burnaby, CA); Yu Feng Liao (Vancouver, CA)
G06T3/4046G06K9/00624G06K9/2054G06K9/6256
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,222,402
App. No.
16/858,234
Granted
Jan 11, 2022
Kind
B2
Abstract

A method of adaptive image enhancement, comprising, receiving a low resolution image, receiving at least one application constraint, detecting at least one scene within the low resolution image, detecting a plurality of regions of interest within the detected at least one scene, prioritizing the detected plurality of regions of interest, ranking the detected plurality of regions of interest based on the prioritization, determining an enhanceable subset of the plurality of regions of interest based on the ranking of the plurality of regions of interest and the at least one application constraint and enhancing the enhanceable subset of the plurality of regions of interest.

Claims (33)

1. A method of adaptive image enhancement, comprising:

receiving a low resolution image from an application, the application having at least one application constraint;

receiving at least one model of the low resolution image;

identifying at least one region of interest within the low resolution image based on the at least one model;

initially prioritizing the at least one region of interest based on the application to an initial priority level; and

enhancing a resolution of the low resolution image based on the identified at least one region of interest and the prioritization of the at least one region of interest based on the at least one application constraint;

reassigning the initial priority level of the identified at least one region of interest to an enhanced test priority level;

determining an enhancement time based on the enhanced test priority level;

determining whether a time constraint of the at least one application constraint is met, if the time constraint is not met, then reducing the enhanced test priority level of a lower initial priority level region of interest until the time constraint is met; and

ranking the at least one region of interest based on the initial priority level wherein the reducing of the enhanced test priority level is based on the ranking of the at least one region of interest.

2. The method of adaptive image enhancement of claim 1 , further comprising:

at least one more region of interest forming a plurality of regions of interest;

spatially mapping the plurality of low resolution images to the plurality of high resolution images;

calculation of residuals for the plurality of low resolution images;

addition of calculated residuals for the plurality of low resolution images to the enhanced subset of the plurality of regions of interest according the mapping to the plurality of high resolution region.

3. The method of adaptive image enhancement of claim 1 , further comprising pre-processing the low resolution image.

4. The method of adaptive image enhancement of claim 1 , further comprising detecting a scene based on the low resolution image and modeling the low resolution image based on the detected scene.

5. The method of adaptive image enhancement of claim 1 , wherein the model of the low resolution image is pre-trained.

6. The method of adaptive image enhancement of claim 5 , wherein the identification of the at least one region of interest is based on the pre-trained model.

7. The method of adaptive image enhancement of claim 1 , wherein the at least one application constraint is based on at least one of a cloud defined policy and a user defined policy.

8. The method of adaptive image enhancement of claim 1 , wherein the resolution enhancement of the low resolution image is hardware based.

9. A method of adaptive image enhancement, comprising:

receiving a low resolution image from an application, the application having a time constraint;

receiving at least one model of the low resolution image;

identifying at least one region of interest within the low resolution image based on the at least one model;

initially prioritizing the at least one region of interest based on the application to an initial priority level;

ranking the at least one region of interest based on the initial priority level;

reassigning the initial priority level of the identified at least one region of interest to an enhanced test priority level;

determining an enhancement time based on the enhanced test priority level;

determining whether the time constraint is met, if the time constraint is not met, then reducing the enhanced test priority level of a lower ranked level region of interest until the time constraint is met; and

enhancing a resolution of the low resolution image based on the ranking of the at least one region of interest.

10. The method of adaptive image enhancement of claim 9 , further comprising detecting a scene based on the low resolution image and modeling the low resolution image based on the detected scene, wherein the model of the low resolution image is pre-trained.

11. The method of adaptive image enhancement of claim 9 , wherein the identification of the at least one region of interest is based on a neural network model.

Continuity (1)
Related Publication 20210334937A1 · Oct 28, 2021