Image content and quality assurance system and method

• US 8,234,168 B1
• Filed: 04/19/2012
• Issued: 07/31/2012
• Est. Priority Date: 04/19/2012
• Status: Active Grant

• Current Assignee: Yahoo AD Tech LLC
• Original Assignee: Luminate, Inc. (Apollo Global Management, Inc.)
• Inventors: Everingham, James R., Lagle Ruiz, Kristen, Waterson, Chris
• Primary Examiner(s): HOAR, COLLEEN A
• Application Number:
  US13/450,807
• Time in Patent Office: 103 Days
• Field of Search: 705/14
• US Class Current: 705/14.64
• CPC Class Codes: G06Q 30/0267 Wireless devices

Abstract

Disclosed herein are computer-implement systems and methods for identifying and analyzing content (e.g., text, images, videos, etc.) published on digital content platforms (e.g., webpages, mobile applications, etc.). Such analysis is then used to determine whether the published content is appropriate for association with (or “hosting of”) a third-party'"'"'s content. In one embodiment, for example, the systems and methods presented are particularly useful for determining the appropriateness of an image published on a digital content platform, prior to providing an advertisement proximate the image. As such, merchants can avoid associating their advertised products/services with vulgar, obscene, or otherwise inappropriate images that may have a negative impact on their brand or reputation.

Claims

1. A computer-implemented method for a service provider to identify images eligible for hosting advertisements on a digital content platform, the method comprising:

(a) using a computer-implemented image collection system to

(1) identify an image published on a digital content platform, and

(2) identify a publisher of the image,

(b) using a computer-implemented image analysis system to

(1) calculate an image hash of the image,

(2) compare the image hash against a database of known ineligible images, wherein if the image hash matches an image hash of a known ineligible image, then flagging the image as ineligible for hosting advertisements,

(3) perform a pattern scan of the image,

(4) compare the pattern scan of the image against a database of known ineligible images, wherein if the pattern scan of the image matches a pattern scan of a known ineligible image, then flagging the image as ineligible for hosting advertisements,

(5) perform a text association analysis of the image, and

(6) compare the text association analysis of the image against a database of known ineligible images, wherein if the text association analysis of the image matches a known ineligible image, then flagging the image as ineligible for hosting advertisements; and

(c) if the image analysis system of step (b) does not flag the image as ineligible for hosting advertisements, then the computer-implemented image analysis system automatically submits the image to an image-advertisement matching engine, wherein the image is linked to an advertisement, and wherein the advertisement is sent to the digital content platform for publication proximate to the image.

2. A computer-implemented method for a service provider to identify images eligible for hosting advertisements on a digital content platform, the method comprising:

(a) using a computer-implemented image collection system to

(1) identify an image published on a digital content platform, and

(2) identify a publisher of the image;

(b) using a computer-implemented image analysis system to assess content within the image, wherein if the image analysis system determines that the image is eligible for hosting advertisements, then the image is automatically submitted to an image-advertisement matching engine, wherein the image is linked to an advertisement, and wherein the advertisement is sent to the digital content platform for publication proximate to the image, and if the image analysis system determines that the image is ineligible for hosting advertisements, then the image is flagged as ineligible and is automatically submitted for relationship test analysis in step (c); and

(c) performing a relationship test analysis by assessing whether the image violates a pre-defined service provider terms of service.

3. The method of claim 2, the method further comprising:

maintaining a publisher trust score; and

updating the publisher trust score for the publisher identified in step (a)(2) based on the image analysis of step (b) or the relationship test analysis of step (c).

4. The method of claim 2, wherein prior to step (b), the method further comprises:

using a computer-implemented processing device to

(1) calculate an image hash of the image, and

(2) compare the image hash against a database of known ineligible images, wherein if the image hash matches an image hash of a known ineligible image, then flagging the image as ineligible for hosting advertisements, and if the image hash does not match an image hash of a known ineligible image, then submitting the image for image analysis in step (b).

5. The method of claim 2, wherein the image analysis system includes a computerized crowdsource interface.

6. The method of claim 5, wherein step (b) is performed by displaying the image to a crowdsource network via the computerized crowdsource interface.

7. The method of claim 2, wherein if the relationship test analysis of step (c) determines that the image does not violate the pre-defined service provider terms of service, the method further comprises:

removing the image flag applied in step (b); and

submitting the image to the image-advertisement matching engine, wherein the image is linked to an advertisement, and wherein the advertisement is sent to the digital content platform for publication proximate to the image.

8. The method of claim 2, wherein if the relationship test analysis of step (c) determines that the image does violate the pre-defined publisher terms of service, the method further comprises:

flagging the publisher identified in step (a) (2) as inappropriate for hosting advertisements.

9. The method of claim 2, wherein prior to step (b), the method further comprises:

using a computer-implemented processing device to

(1) perform a pattern scan of the image, and

(2) compare the pattern scan of the image against a database of known ineligible images, wherein if the pattern scan of the image matches a pattern scan of a known ineligible image, then flagging the image as ineligible for hosting advertisements, and if the pattern scan of the image does not match a pattern scan of a known ineligible image, then submitting the image for image analysis in step (b).

10. The method of claim 2, wherein prior to step (b), the method further comprises:

using a computer-implemented processing device to

(1) perform a text association analysis of the image, and

(2) compare the text association analysis of the image against a database of known ineligible images, wherein if the text association analysis of the image matches a known ineligible image, then flagging the image as ineligible for hosting advertisements, and if the text association analysis of the image does not match a known ineligible image, then submitting the image for image analysis in step (b).

11. A computer-implemented method for a service provider to match images published on a digital content platform with contextually relevant advertisements, the method comprising:

(a) using a computer-implemented image collection system to identify an image published on a digital content platform;

(b) using a computer-implemented image analysis system to assess content within the image, wherein the image analysis system determines whether the image is eligible for hosting advertisements based on the content within the image; and

if the image is eligible for hosting advertisements

(c) using an image-advertisement matching engine to link a contextually relevant advertisement to the image; and

(d) providing the contextually relevant advertisement to the digital content platform for publication proximate to the image.

12. The method of claim 11, wherein if the image is not eligible for hosting advertisements, the method comprises:

flagging the image as ineligible for hosting advertisements;

automatically submitting the image for relationship test analysis; and

performing a relationship test analysis by assessing whether the image violates pre-defined terms.

13. The method of claim 11, further comprising:

using the computer-implemented image collection system to identify a publisher of the image.

14. The method of claim 13, further comprising:

maintaining a publisher trust score; and

updating the publisher trust score based on the image analysis of step (b).

15. The method of claim 11, wherein the digital content platform is a website.

16. The method of claim 11, wherein the digital content platform is a mobile application.

17. The method of claim 11, wherein the digital content platform is a social network.

18. The method of claim 11, further comprising:

using a computer-implemented processing device to calculate an image hash of the image; and

using the computer-implemented processing device to compare the image hash against a database of known ineligible images, wherein if the image hash matches an image hash of a known ineligible image, then flagging the image as ineligible for hosting advertisements.

19. The method of claim 11, wherein the image analysis system includes a computerized crowdsource interface.

20. The method of claim 19, wherein step (b) is performed by displaying the image to a crowdsource network via the computerized crowdsource interface.

21. The method of claim 19, wherein step (c) is performed by displaying the image to a crowdsource network via the computerized crowdsource interface.

22. A computer-readable storage medium, comprising:

instructions executable by at least one processing device that, when executed, cause the processing device to;

(a) identify an image published on a digital content platform;

(b) identify a publisher of the image;

(c) assess content within the image to determine whether the image is eligible for hosting advertisements;

(d) link a contextually relevant advertisement to the image, if the image is eligible for hosting advertisements; and

(d) provide the contextually relevant advertisement to the digital content platform for publication proximate to the image.