Granted Patent B2

US 12,423,938 · App. 17/872,932 · Granted Sep 23, 2025

Inventors: Ofir Ezrielev (Be'er Sheva, IL); Amihai Savir (Newton, MA); Oshry Ben-Harush (Cedar Park, TX)

Assignee: Dell Products L.P.

G06V10/25G06V10/235G06V10/764

Abstract

Claims (76)

1. A method for managing images for use by downstream consumers, the method comprising:

obtaining an image of the images;

obtaining a primary area of interest for the image;

while obtaining the primary area of interest, obtaining focus indicators for the image based on a process through which the primary area of interest is obtained;

obtaining, based on the focus indicators, an auxiliary area of interest for the image, the primary area of interest and the auxiliary area of interest both being areas of interest within the image that are separate from areas of no interest within the image;

classifying the auxiliary area of interest as being relevant or irrelevant to the primary area of interest to obtain a relevant auxiliary area of interest or an irrelevant auxiliary of interest, respectively, the relevant auxiliary area of interest and the irrelevant auxiliary area of interest being different from the areas of no interest;

segmenting the image into image segments comprising at least:

non-mixed image segments containing only one of the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

mixed image segments containing a mix of at least two among the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest; and

storing the image segments of the image in storage tiers based on an amount of the primary area of interest, the relevant auxiliary area of interest, the irrelevant auxiliary area of interest, and the areas of no interest contained within each of the image segments to facilitate use of the stored image by the downstream consumers, wherein each of the mixed image segments is stored into the storage tiers based on a likelihood of each of the mixed image segments being accessed after being stored in the storage tiers, the likelihood being based on an access likelihood hierarchy established between the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

wherein, for a mixed image segment of the mixed image segments, the mixed image segment is stored into the storage tiers based on a highest access likelihood portion contained within the mixed image segment, the highest access likelihood portion being determined based on a content of the mixed image segment and the access likelihood hierarchy.

2. The method of claim 1 , wherein obtaining the focus indicators comprises:

while a subject matter expert views the image to annotate the primary area of interest for the image, correlating the subject matter expert's view of the image to portions of the image during the process through which the primary area of interest is obtained to obtain the focus indicators.

3. The method of claim 2 , wherein the auxiliary area of interest is a diagnostically relevant area of interest in an instance of the focus indicators that specify that the subject matter expert focused on a portion of the portions of the image that covers only the auxiliary area of interest instead of an entirety of the image for a duration of time that exceeded a first time threshold but that is still less than a second time threshold associated with the primary area of interest.

4. The method of claim 2 , wherein the auxiliary area of interest is a diagnostically irrelevant area of interest in an instance of the focus indicators that specify that the subject matter expert focused on a portion of the portions of the image for a duration of time that failed to meet a time threshold associated with the primary area of interest.

5. The method of claim 1 , wherein obtaining the primary area of interest comprises:

presenting portions of an image to a subject matter expert;

receiving user input from the subject matter expert; and

establishing the primary area of interest based on the user input.

6. The method of claim 5 , wherein obtaining the focus indicators comprises:

monitoring interest levels of the subject matter expert for the portions of the image; and

establishing the focus indicators based on the interest levels.

7. The method of claim 6 , wherein monitoring the interest levels comprises:

tracking eye movement of the subject matter expert during presentation of the portions of the image.

8. The method of claim 7 , further comprising:

establishing an interest hierarchy for the portions of the image based on the tracked eye movement.

9. The method of claim 1 , wherein the auxiliary area of interest is stored in a second tier of tiered storages when classified as being relevant to the primary interest and in a fourth tier different when classified as being irrelevant to the primary interest, the second tier and the fourth tier being different from a first tier of the tiered storage adapted for storing the primary area of interest.

10. The method of claim 1 , wherein the image segments are classified into four segment categories comprising a high area of interest segment that comprises the primary area of interest, an irrelevant area of interest segment that comprises the irrelevant auxiliary area of interest, a low area of interest segment that comprises the relevant auxiliary area of interest, and a no area of interest segment that comprises the areas of no interest.

11. The method of claim 1 , further comprising:

after storing the image segments of the image in the storage tiers, obtaining a request to view the image; and

making, in response to the request, only the image segments, among the mixed image segments and the non-mixed image segments, that comprise the primary area of interest, the relevant auxiliary area of interest, and the irrelevant auxiliary area of interest available to be displayed on a graphical user interface without providing access to an entirety of the image.

12. The method of claim 11 , wherein obtaining the primary area of interest and the auxiliary area of interest for the image comprises:

generating a heat map for the image, the heat map comprising values for each pixel of pixels making up the image, and the values being weighted based on a duration of display of each of the pixels on a graphical user interface on which the image is displayed; and

filtering the heat map using a plurality of thresholds to generate at least one interest indicator among one or more interest indicators for one or more portions of the heat map, each of the one or more interest indicators corresponds to one of the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest.

13. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations for managing images for use by downstream consumers, the operations comprising:

obtaining an image of the images;

obtaining a primary area of interest for the image;

while obtaining the primary area of interest, obtaining focus indicators for the image based on a process through which the primary area of interest is obtained;

obtaining, based on the focus indicators, an auxiliary area of interest for the image, the primary area of interest and the auxiliary area of interest both being areas of interest within the image that are separate from areas of no interest within the image;

classifying the auxiliary area of interest as being relevant or irrelevant to the primary area of interest to obtain a relevant auxiliary area of interest or an irrelevant auxiliary of interest, respectively, the relevant auxiliary area of interest and the irrelevant auxiliary area of interest being different from the areas of no interest;

segmenting the image into image segments comprising at least:

non-mixed image segments containing only one of the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

mixed image segments containing a mix of at least two among the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest; and

storing the image segments of the image in storage tiers based on an amount of the primary area of interest, the relevant auxiliary area of interest, the irrelevant auxiliary area of interest, and the areas of no interest contained within each of the image segments to facilitate use of the stored image by the downstream consumers, wherein each of the mixed image segments is stored into the storage tiers based on a likelihood of each of the mixed image segments being accessed after being stored in the storage tiers, the likelihood being based on an access likelihood hierarchy established between the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

wherein, for a mixed image segment of the mixed image segments, the mixed image segment is stored into the storage tiers based on a highest access likelihood portion contained within the mixed image segment, the highest access likelihood portion being determined based on a content of the mixed image segment and the access likelihood hierarchy.

14. The non-transitory machine-readable medium of claim 13 , wherein obtaining the primary area of interest comprises:

presenting portions of an image to a subject matter expert;

receiving user input from the subject matter expert; and

establishing the primary area of interest based on the user input.

15. The non-transitory machine-readable medium of claim 14 , wherein obtaining the focus indicators comprises:

monitoring interest levels of the subject matter expert for the portions of the image; and

establishing the focus indicators based on the interest levels.

16. The non-transitory machine-readable medium of claim 15 , wherein monitoring the interest levels comprises:

tracking eye movement of the subject matter expert during presentation of the portions of the image.

17. A data processing system, comprising:

a processor; and

a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations for managing images for use by downstream consumers, the operations comprising:

obtaining an image of the images;

obtaining a primary area of interest for the image;

while obtaining the primary area of interest, obtaining focus indicators for the image based on a process through which the primary area of interest is obtained;

obtaining, based on the focus indicators, an auxiliary area of interest for the image, the primary area of interest and the auxiliary area of interest both being areas of interest within the image that are separate from areas of no interest within the image;

classifying the auxiliary area of interest as being relevant or irrelevant to the primary area of interest to obtain a relevant auxiliary area of interest or an irrelevant auxiliary of interest, respectively, the relevant auxiliary area of interest and the irrelevant auxiliary area of interest being different from the areas of no interest;

segmenting the image into image segments comprising at least:

non-mixed image segments containing only one of the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

mixed image segments containing a mix of at least two among the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest; and

storing the image segments of the image in storage tiers based on an amount of the primary area of interest, the relevant auxiliary area of interest, the irrelevant auxiliary area of interest, and the areas of no interest contained within each of the image segments to facilitate use of the stored image by the downstream consumers, wherein each of the mixed image segments is stored into the storage tiers based on a likelihood of each of the mixed image segments being accessed after being stored in the storage tiers, the likelihood being based on an access likelihood hierarchy established between the primary area of interest, the relevant auxiliary area of interest, irrelevant auxiliary area of interest, and the areas of no interest, and

wherein, for a mixed image segment of the mixed image segments, the mixed image segment is stored into the storage tiers based on a highest access likelihood portion contained within the mixed image segment, the highest access likelihood portion being determined based on a content of the mixed image segment and the access likelihood hierarchy.

18. The data processing system of claim 17 , wherein obtaining the primary area of interest comprises:

presenting portions of an image to a subject matter expert;

receiving user input from the subject matter expert; and

establishing the primary area of interest based on the user input.

19. The data processing system of claim 18 , wherein obtaining the focus indicators comprises:

monitoring interest levels of the subject matter expert for the portions of the image; and

establishing the focus indicators based on the interest levels.

20. The data processing system of claim 19 , wherein monitoring the interest levels comprises:

tracking eye movement of the subject matter expert during presentation of the portions of the image.

Assignments (1)

Continuity (1)

References Cited (120)