Image anonymization using analytics tool

• US 9,858,696 B2
• Filed: 09/18/2015
• Issued: 01/02/2018
• Est. Priority Date: 09/18/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Gordon, Itai, Hagelund, Peter, Henderson, David J.
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US14/858,605
• Publication Number:
  US 20170084065A1
• Time in Patent Office: 837 Days
• Field of Search: 382190, 382195, 382215
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text

Abstract

A mechanism is provided in a data processing system for image anonymization. The mechanism identifies a plurality of items in an image and extracts properties of the plurality of items into one or more data structures. The mechanism queries an analytics tool about whether a combination of properties of the plurality of items identifies a person in the image. Responsive to determining a given combination of properties of the plurality of items identifies a person in the image, the mechanism modifies the plurality of items in the image to form an anonymized image and outputs the anonymized image.

Claims

1. A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on a computing device, causes the computing device to:

identify a plurality of items in an image;

perform image analysis to identify properties of the plurality of items;

store the properties of the plurality of items into one or more data structures;

query an analytics tool to determine, for each combination of two or more properties of the plurality of items, a probability that the combination of two or more properties of the plurality of items identifies a person in the image;

determine that a given combination of two or more properties of the plurality of items identifies a person in the image by comparing its corresponding probability to a threshold;

responsive to determining the given combination of two or more properties of the plurality of items identifies a person in the image, modify the plurality of items in the image to form an anonymized image; and

output the anonymized image.

2. The computer program product of claim 1, wherein identifying the plurality of items comprises performing image processing on the image.

3. The computer program product of claim 1, wherein a given item within the plurality of items comprises a person and wherein properties of the given item comprise at least one of hair color, eye color, body type, height, clothing items, clothing color, or clothing style.

4. The computer program product of claim 1, wherein a given item within the plurality of items comprises a vehicle and wherein properties of the given item comprise at least one of make, model, color, vehicle type, or state of license plate.

5. The computer program product of claim 1, wherein a given item within the plurality of items comprises a dwelling and wherein properties of the given item comprise at least one of dwelling type or signage.

6. The computer program product of claim 1, wherein querying the analytics tool comprises identifying a plurality of combinations of properties and generating a query for the analytics tool for each of the plurality of combinations of properties.

7. The computer program product of claim 6, wherein identifying the plurality of combinations of properties comprises starting with single property sets and proceeding with gradually larger property sets.

8. The computer program product of claim 1, wherein the analytics tool whether a given combination of properties of the plurality of items identifies a given person in the image based on a plurality of data sources.

9. The computer program product of claim 8, wherein the plurality of data sources comprise at least one of weather data, demographic data, sales statistics, or user rights and roles.

10. The computer program product of claim 1, wherein modifying the plurality of items in the image comprises replacing, obscuring, or obfuscating one or more items corresponding to the given combination of two or more properties.

11. An apparatus comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to;

identify a plurality of items in an image;

perform image analysis to identify properties of the plurality of items;

store the properties of the plurality of items into one or more data structures;

query an analytics tool to determine, for each combination of two or more properties of the plurality of items, a probability that the combination of two or more properties of the plurality of items identifies a person in the image;

determine that a given combination of two or more properties of the plurality of items identifies a person in the image by comparing its corresponding probability to a threshold;

responsive to determining a given combination of properties of the plurality of items identifies a person in the image, modify the plurality of items in the image to form an anonymized image; and

output the anonymized image.

12. The apparatus of claim 11, wherein identifying the plurality of items comprises performing image processing on the image.

13. The apparatus of claim 11, wherein a given item within the plurality of items comprises a person and wherein properties of the given item comprise at least one of hair color, eye color, body type, height, clothing items, clothing color, or clothing style.

14. The apparatus of claim 11, wherein a given item within the plurality of items comprises a vehicle and wherein properties of the given item comprise at least one of make, model, color, vehicle type, or state of license plate.

15. The apparatus of claim 11, wherein a given item within the plurality of items comprises a dwelling and wherein properties of the given item comprise at least one of dwelling type or signage.

16. The apparatus of claim 11, wherein querying the analytics tool comprises identifying a plurality of combinations of properties and generating a query for the analytics tool for each of the plurality of combinations of properties.

17. The apparatus of claim 16, wherein identifying the plurality of combinations of properties comprises starting with single property sets and proceeding with gradually larger property sets.

18. The apparatus of claim 11, wherein the analytics tool whether a given combination of properties of the plurality of items identifies a given person in the image based on a plurality of data sources.

19. The apparatus of claim 18, wherein the plurality of data sources comprise at least one of weather data, demographic data, sales statistics, or user rights and roles.

20. The apparatus of claim 11, wherein modifying the plurality of items in the image comprises replacing, obscuring, or obfuscating one or more items corresponding to the given combination of two or more properties.