Granted Patent B2

US 12,416,062 · App. 18/344,337 · Granted Sep 16, 2025

Inventors: David Watola (Irvine, CA); Jay E. Cormier (Laguna Niguel, CA); Brian Kim (San Clemente, CA)

Assignee: Eyedaptic, Inc.

C22B1/00A61F9/08C22B11/00C22B15/00G02B27/0101G06F3/013G06V20/10G09B21/008G02B2027/0138G02B2027/014G02B2027/0178

Abstract

Claims (64)

1. A method of providing enhanced vision for a low-vision user, comprising the steps of:

receiving real-time video images with a visual aid device;

identifying, with the visual aid device, a subset of each of the real-time video images corresponding to a focus of attention of the low-vision user;

extracting, with the visual aid device, structures of interest within the subset of each of the real-time video images;

examining the extracted structures of interest with the visual aid device; and

discarding structures of interest that include parameters that are outside of pre-selected bounds identifying, with the visual aid device, at least one group of structures of interest that are organized in a row or column;

determining, with the visual aid device, a dimension of the at least one group of structures of interest;

adjusting, with the visual aid device, a magnification of the at least one group of structures of interest to match a preferred structure dimension;

forming, with the visual aid device, an enhanced video stream that includes the real-time images and the at least one group of structures of interest with adjusted magnification; and

displaying the enhanced video stream on a display of the visual aid device.

2. The method of claim 1 , wherein the subset comprises a fixed area of pixels within each of the real-time video images.

3. The method of claim 2 , wherein the fixed area of pixels is at a central portion of each of the real-time video images.

4. The method of claim 1 , wherein the structures of interest comprise text-like structures.

5. The method of claim 1 , wherein the structures of interest comprise facial structures.

6. The method of claim 1 , wherein the structures of interest are selected from the group consisting of letters, numerals, pictograms, glyphs and icons.

7. The method of claim 1 , wherein the visual aid device automatically identifies the subset of each of the real-time video images based on behavioral trends of a user of the visual aid device.

8. The method of claim 7 , wherein behavioral trends of the user comprise a user history of input control sequences and a timing of these input control sequences.

9. The method of claim 1 , wherein the visual aid device automatically identifies the subset of each of the real-time video images based on the contents of previously processed video images.

10. The method of claim 1 , further comprising:

tracking a gaze of the low-vision user with the visual aid device; and

wherein the visual aid device automatically identifies the subset of each of the real-time video images based on the gaze of the low-vision user.

11. The method of claim 1 , further comprising preprocessing the subset of real-time video images to reduce noise or accommodate other unwanted interference patterns.

12. The method of claim 11 , where in the preprocessing comprises enhancing a contrast component of the subset of real-time video images.

13. The method of claim 11 , wherein the preprocessing comprises increasing a sharpness component of the subset of real-time video images.

14. The method of claim 11 , wherein the preprocessing comprises increasing a detail component of the subset of real-time video images.

15. The method of claim 1 , wherein the extracting step further comprises extracting maximally stable extremal regions from the subset of real-time video images.

16. The method of claim 1 , wherein the pre-selected bounds include a length threshold parameter.

17. The method of claim 1 , wherein the pre-selected bounds include a width threshold parameter.

18. The method of claim 1 , wherein the pre-selected bounds include a pixel-count threshold parameter.

19. The method of claim 1 , wherein the pre-selected bounds include an aspect ratio threshold parameter.

20. The method of claim 1 , wherein identifying the at least one group of structures of interest that are organized in a row or column further comprises:

modeling a probability density function of vertical centers for the structures of interest, wherein the probability density function includes one or more pdf peaks;

identifying a maximum peak from the pdf peaks of the probability density functions; and

retaining only structures of interest that are statistically likely to be associated with the maximum peak.

21. The method of claim 20 , wherein structures of interest that are statistically likely to be associated with the maximum peak comprises structures of interest that are consistent in size.

22. The method of claim 1 , further comprising determining, with the visual aid device, a statistical confidence that the at least one group of structures of interest are organized in a row or column.

23. The method of claim 22 , wherein adjusting a magnification of at least one group of structures of interest further comprises adjusting a magnification of the at least one group of structures of interest to match the preferred structure size when the statistical confidence is above a confidence threshold.

24. The method of claim 1 , wherein the receiving step further comprises:

obtaining real-time video images with a camera of the visual aid device; and

receiving the real-time video images from the camera with a visual aid device.

25. The method of claim 24 , further comprising:

sensing a motion component of the visual aid device as it obtains the real-time video images;

determining a motion state of the visual aid device based on the sensed motion component; and

wherein the adjusting a magnification step, the forming an enhanced video stream step, and the displaying the enhanced video stream step are performed only if the motion state comprises a focused attention state.

26. A method of providing enhanced vision for a low-vision user, comprising the steps of:

receiving real-time video images with a visual aid device;

identifying, with the visual aid device, a subset of each of the real-time video images corresponding to a focus of attention of the low-vision user;

extracting, with the visual aid device, structures of interest within the subset of each of the real-time video images;

determining, with the visual aid device, a dimension of the at least one group of structures of interest;

determining, with the visual aid device, a statistical confidence that the at least one group of structures of interest are organized in a row or column;

adjusting, with the visual aid device, a magnification of the at least one group of structures of interest to match a preferred structure dimension, wherein adjusting a magnification of at least one group of structures of interest further comprises adjusting a magnification of the at least one group of structures of interest to match the preferred structure size when the statistical confidence is above a confidence threshold;

forming, with the visual aid device, an enhanced video stream that includes the real-time images and the at least one group of structures of interest with adjusted magnification; and

displaying the enhanced video stream on a display of the visual aid device.

27. A method of providing enhanced vision for a low-vision user, comprising the steps of:

obtaining real-time video images with a camera of a visual aid device;

sensing a motion component of the visual aid device as it obtains the real-time video images;

determining a motion state of the visual aid device based on the sensed motion component;

receiving the real-time video images from the camera with a visual aid device;

identifying, with the visual aid device, a subset of each of the real-time video images corresponding to a focus of attention of the low-vision user;

extracting, with the visual aid device, structures of interest within the subset of each of the real-time video images;

determining, with the visual aid device, a dimension of the at least one group of structures of interest;

adjusting, with the visual aid device, a magnification of the at least one group of structures of interest to match a preferred structure dimension only if the motion state comprises a focused attention state;

forming, with the visual aid device, an enhanced video stream that includes the real-time images and the at least one group of structures of interest with adjusted magnification; and

displaying the enhanced video stream on a display of the visual aid device only if the motion state comprises a focused attention state.

Assignments (1)

Continuity (3)

References Cited (171)