METHOD AND SYSTEM FOR FACILITATING RETRIEVAL OF INFORMATION FROM IMAGES

• US 20140321775A1
• Filed: 04/25/2013
• Published: 10/30/2014
• Est. Priority Date: 04/25/2013
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Wu, Wencheng, Dalal, Edul N.
• Granted Patent: US 9,218,579 B2
• Time in Patent Office: Days
• US Class Current: 382/308
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06V 10/95 structured as a network, e.g. client-server architectures
• CPC Class Codes: G06V 10/987 with the intervention of an operator
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

Disclosed embodiments relate to systems and methods for facilitating retrieval of information from an image. The one or more image characteristics are modified to degrade the original image. The degraded image is provided as a first task to at least one first remote worker, the first task corresponds to an activity of tagging one or more regions of interest in the degraded image by the at least one first remote worker. Further, one or more second images are generated based on the one or more tagged regions of interest and are provided as one or more second tasks to at least one second remote worker, the one or more second tasks correspond to an activity of retrieving information from the one or more second images.

Claims

1. A method for retrieving information from an original image, the method comprising:

modifying one or more image characteristics to degrade the original image;

providing the degraded image as a first task to at least one first remote worker or a crowdsourcing platform server, wherein the first task corresponds to an activity of tagging one or more regions of interest in the degraded image by the at least one first remote worker;

receiving the one or more tagged regions of interest in response to the first task; and

providing one or more second images as one or more second tasks to at least one second remote worker or the crowdsourcing platform server, wherein the one or more second images are generated based on the one or more tagged regions of interest, and wherein the one or more second tasks correspond to an activity of retrieving the information from the one or more second images,

wherein the method is performed by one or more processors.

2. The method of claim 1 further comprising cropping one or more portions corresponding to the one or more tagged regions of interest from the original image to generate the one or more second images.

3. The method of claim 1 further comprising modifying the one or more image characteristics of the one or more tagged regions of interest to generate the one or more second images.

4. The method of claim 1 further comprising determining a degradation level of the first task based on at least one of a response time for the first task, a comparison between the responses for the multiple first tasks, a comparison between the response for the first task and the information in the original image, or a feedback from the at least one first remote worker.

5. The method of claim 4 further comprising reducing the degradation of the degraded image based on the determined degradation level.

6. The method of claim 1, wherein the information in the degraded image is not recognizable, and wherein the location of the information in degraded image is identifiable.

7. The method of claim 1, wherein the modifying the one or more image characteristics comprises performing at least one of a color to gray conversion of the original image, a spatial resolution reduction of the original image, an addition of noise in the original image, or a mapping of a color in the original image to a random color based on a pre-determined rule.

8. The method of claim 1, wherein the one or more second tasks correspond to at least one of a license plate recognition, a handwriting recognition in a digitized document, or a video labeling in a traffic surveillance system.

9. The method of claim 1, wherein the image corresponds to at least one of a digitized document having handwritten content, a still image, or a frame of a video footage.

10. A system for facilitating retrieval of information from an original image, the system comprising:

a memory comprising;

an image-processing module configured to modify one or more image characteristics to degrade the original image, and

a communication manager configured to;

provide the degraded image as a first task to at least one first remote worker or a crowdsourcing platform server, wherein the first task corresponds to an activity of tagging one or more regions of interest in the degraded image by the at least one first remote worker,

receive the one or more tagged regions of interest in response to the first task, and

provide one or more second images as one or more second tasks to at least one second remote worker or the crowdsourcing platform server, wherein the one or more second images are generated based on the one or more tagged regions of the interest, and wherein the one or more second tasks correspond to an activity of retrieving the information from the one or more second images; and

one or more processors configured to execute the communication manager, the feature determination module, and the statistical model.

11. The system of claim 10, wherein the image-processing module is further configured to crop one or more portions corresponding to the one or more tagged regions of interest from the original image to generate the one or more second images.

12. The system of claim 11 further comprising a task creation module configured to create the first task and the one or more second tasks.

13. The system of claim 10, wherein the image corresponds to at least one of a digitized document having handwritten content, a still image, or a frame of a video footage.

14. The system of claim 10, wherein the image-processing module is configured to modify the one or more image characteristics by at least one of a color to gray conversion of the original image, a spatial resolution reduction of the original image, an addition of noise in the image, or a mapping of a color in the original image to a random color based on a pre-determined rule.

15. The system of claim 10, wherein the one or more second tasks correspond to at least one of a license plate recognition, a handwriting recognition in a digitized document, or a video labeling in traffic surveillance system.

16. The system of claim 10, wherein the image-processing module is further configured to determine a degradation level of the first task based on at least one of a response time for the first task, a comparison of the responses for multiple first tasks, a comparison of the response for the first task and the information in the original image, or a feedback from the at least one first remote worker.

17. The system of claim 16, wherein the image-processing module is further configured to reduce the degradation of the degraded image based on the determined degradation level.

18. The system of claim 10, wherein the information in the degraded image is not recognizable, and wherein the location of the information in the degraded image is identifiable.

19. A computer program product for use with a computer, the computer program product comprising a non-transitory computer readable medium, the non-transitory computer readable medium stores a computer program code for facilitating retrieval of information from an original image, the computer program code performing a method, the method comprising:

modifying one or more image characteristics to degrade the original image;

providing the degraded image as a first task to at least one first remote worker or a crowdsourcing platform server, wherein the first task corresponds to an activity of tagging one or more regions of interest in the degraded image by the at least one first remote worker;

receiving the one or more tagged regions of interest in response to the first task; and

providing one or more second images as one or more second tasks to at least one second remote worker or the crowdsourcing platform server, wherein the one or more second images are generated based on the one or more tagged regions of interest, and wherein the one or more second tasks correspond to an activity of retrieving the information from the one or more second images.

20. The computer program product of claim 19, wherein the method further comprises cropping one or more portions corresponding to the one or more tagged regions of interest from the original image to generate the one or more second images.