Adaptive thresholding for image recognition

• US 9,055,384 B2
• Filed: 10/04/2013
• Issued: 06/09/2015
• Est. Priority Date: 06/07/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Ivanchenko, Volodymyr V., Heller, Geoffrey Scott, Suplee, Richard Howard III, Bibireata, Daniel
• Primary Examiner(s): Do, Ahn
• Application Number:
  US14/046,814
• Publication Number:
  US 20140187223A1
• Time in Patent Office: 613 Days
• Field of Search: 382/176, 382/173, 382/190, 382/141, 382/194, 382/199, 382/305, 1/1, 345/172, 345/173, 379/88.01, 379/88.14, 379/88.18, 379/265.09, 707/E17.008, 707/999.03, 704/231, 715/212, 715/809
• US Class Current: 1/1
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names
• CPC Class Codes: G06V 30/1478 of characters or characters lines
• CPC Class Codes: H04W 4/50 Service provisioning or reconfiguring

Abstract

Various approaches for providing textual information to an application, system, or service are disclosed. In particular, various embodiments enable a user to capture an image with a camera of a portable computing device. The computing device is capable of taking the image and processing it to recognize, identify, and/or isolate the text in order to forward the text to an application or function. The application or function can then utilize the text to perform an action in substantially real-time. The text may include an email, phone number, URL, an address, and the like and the application or function may be dialing the phone number, navigating to the URL, opening an address book to save contact information, displaying a map to show the address, and so on. Adaptive thresholding can be used to account for variations across an image, in order to improve the accuracy and efficiency of text recognition processes.

Claims

1. A computer-implemented method comprising:

obtaining at least one image of information associated with an event captured by a camera of a portable computing device;

analyzing the at least one image to locate a region of the image having visual properties identifying the event;

processing the region using an optical recognition algorithm to identify the event based on the visual properties;

determining at least one action available to a user, the at least one action being related to the event; and

causing the portable computing device to execute the at least one action related to the event.

2. The computer-implemented method of claim 1, wherein an application on the portable computing device executes the at least one action related to the event, the application being at least one of a web browser and the action being associated with navigating to a webpage associated with the event when the visual properties indicate a web address or a map application and the action being associated with displaying location information for the event when the visual properties indicate a physical address.

3. The computer-implemented method of claim 1, wherein analyzing the at least one image includes:

analyzing pixel values adjacent one or more potential edge locations;

determining, based at least in part on the pixel values, a character pixel value threshold;

determining a set of pixels having a pixel value at least meeting the character pixel value threshold; and

assigning the set of pixels as a character portion.

4. The computer-implemented method of claim 3, wherein analyzing the pixel values includes analyzing a first set of pixels on a first side of each potential edge location and analyzing a second set of pixels on a second side of each potential edge location.

5. A computer-implemented method comprising:

under control of one or more computer systems configured with executable instructions,

obtaining image information using a camera of a computing device;

identifying an event associated with the image information;

analyzing text associated with the event using at least one of a computer vision algorithm or an optical character recognition algorithm to identify the text in the image information;

determining at least one action available to a user, the action being related to the event; and

causing the computing device to perform the at least one action related to the event.

6. The computer-implemented method of claim 5, wherein the at least one action is at least one of dialing a phone number, opening an email application for composing an email, navigating to a URL for the event, or displaying a map to show an address associated with the event.

7. The computer-implemented method of claim 5, further comprising:

capturing, using the camera, the image information of an advertisement of the event, the image information being captured automatically in response to the camera being focused on the advertisement, in response to a user input, or combination thereof.

8. The computer-implemented method of claim 7, further comprising:

identifying text in the image information of the advertisement; and

determining a pattern associated with at least one of a pattern indicative of a telephone number for event information, a pattern indicative of an email address associated with the event, a pattern indicative of a URL, or a pattern indicative of a street address.

9. The computer-implemented method of claim 5, further comprising:

analyzing, for each of a plurality of regions, the image information to determine a character pixel value threshold.

10. The computer-implemented method of claim 9, wherein analyzing the image information includes:

analyzing first pixels on a first side of an edge location; and

analyzing second pixels on a second side of the edge location to determine the character pixel value threshold.

11. The computer-implemented method of claim 5, wherein identifying the event includes recognizing text in the image information by at least:

identifying, in the image information, an edge location;

determining a character pixel value threshold based at least in part on pixel values associated with the edge location; and

determining a text character portion based at least in part on a pixel value at least meeting the character pixel value threshold.

12. The computer-implemented method of claim 11, further comprising:

determining the edge location based at least in part on a change in pixel intensity or saturation corresponding to a value at least meeting the character pixel value threshold.

13. The computer-implemented method of claim 11, further comprising:

analyzing first pixels on a first side of an edge location; and

analyzing second pixels on a second side of the edge location to identify the at least one edge location.

14. The computer-implemented method of claim 5, further comprising:

prompting a user to confirm intent to perform the at least one function.

15. A computing device, comprising:

a processor; and

a memory device including instructions that, when executed by the processor, cause the processor to;

obtain at least one image of an advertisement associated with an event, the at least one image being captured by a camera of the computing device;

process the at least one image of the advertisement to locate at least one region having properties of text;

analyze the at least one region using an optical character recognition algorithm to recognize the text associated with the event;

identify a text pattern corresponding to the recognized text;

determine an application associated with the text pattern; and

cause the recognized text to be sent to the application for performing an action with the text pattern associated with the event.

16. The computing device of claim 15, wherein the text pattern includes at least one of a pattern indicative of a telephone number for information associated with the event, a pattern indicative of an email address associated with the event, a pattern indicative of a URL associated with the event, or a pattern indicative of a street address for the event.

17. The computing device of claim 15, wherein the instructions that, when executed by the processor, further cause the processor to:

analyze the at least one image to identify a character pixel value threshold for each of a plurality of regions of the at least one image.

18. The computing device of claim 17, further comprising:

analyze first pixels on a first side of an edge location; and

analyze second pixels on a second side of the edge location to determine a background pixel value.

19. The computing device of claim 15, wherein the action is at least one of dialing an event phone number in response to the recognized text indicating a phone number, opening an email application for composing an email in response to the recognized text indicating an email, navigating to a URL for the event in response to the recognized text indicates a web address, or displaying a map to show an address for the event in response to the recognized text indicating a physical address.