System and method of determining building numbers

• US 9,436,886 B2
• Filed: 03/17/2015
• Issued: 09/06/2016
• Est. Priority Date: 07/12/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Wu, Bo, Bissacco, Alessandro, Smith, Raymond W., Cheung, Kong Man, Frome, Andrea, Urbach, Shlomo
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US14/660,141
• Publication Number:
  US 20160188991A1
• Time in Patent Office: 539 Days
• Field of Search: 382/101, 382/177, 382/182, 382/187, 382/229, 382/284, 701/409, 701/436, 701/484, 701/494, 701/532
• US Class Current: 1/1
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06Q 50/10 Services
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks
• CPC Class Codes: G06V 30/424 Postal images, e.g. labels or addresses on parcels or postal envelopes

Abstract

A system and method is provided for automatically recognizing building numbers in street level images. In one aspect, a processor selects a street level image that is likely to be near an address of interest. The processor identifies those portions of the image that are visually similar to street numbers, and then extracts the numeric values of the characters displayed in such portions. If an extracted value corresponds with the building number of the address of interest such as being substantially equal to the address of interest, the extracted value and the image portion are displayed to a human operator. The human operator confirms, by looking at the image portion, whether the image portion appears to be a building number that matches the extracted value. If so, the processor stores a value that associates that building number with the street level image.

Claims

1. A method comprising:

selecting, by one or more processors, an image associated with a street address that includes a building number;

identifying, by the one or more processors, a portion of the image containing characters associated with building numbers;

extracting, by the one or more processors, an alphanumeric value from the characters within the portion of the image;

comparing, by the one or more processors, the alphanumeric value to the building number of the street address;

determining, by the one or more processors, a score based on the comparison of the alphanumeric value to the building number of the street address;

determining whether the score meets or exceeds a first predetermined threshold; and

storing, by the one or more processors, an association between the alphanumeric value and the image when the score meets or exceeds the first predetermined threshold.

2. The method of claim 1, wherein the score is based on at least one of:

confidence that the extracted alphanumeric value matches the characters in the image;

a format of the extracted alphanumeric value;

a number of digits in the extracted alphanumeric value;

a count of images in which the extracted alphanumeric value appears;

or a difference between an expected address and the extracted alphanumeric value.

3. The method of claim 1, further comprising determining, by the one or more processors, whether the score is at or below a second predetermined threshold, wherein the association between the alphanumeric value and the building number is not stored if the score is at or below the second predetermined threshold.

4. The method of claim 3, further comprising requesting confirmation when the score is below the first predetermined threshold and above the second predetermined threshold.

5. The method of claim 1 wherein identifying a portion of the image containing characters comprises comparing a plurality of portions of the image to image patterns of building numbers.

6. The method of claim 1 wherein comparing the alphanumeric value to the building number comprises comparing the alphanumeric value to a set of values that includes the building number of the street address.

7. The method of claim 1 wherein comparing the alphanumeric value to the building number comprises determining whether the alphanumeric value comprises a numeric value that is equivalent to the building number of the street address.

8. A system comprising:

a memory operative to store a street level image; and

one or more processors in communication with the memory, the one or more processors being configured to;

identify a portion of the street level image such that the portion contains characters associated with building numbers;

extract an alphanumeric value representing the characters within the portion;

compare the alphanumeric value to a range of building numbers that are expected to be near the geographical location at which the street level image was captured;

determine a score based on the comparison of the alphanumeric value to the building number of the street address;

determine whether the score meets or exceeds a first predetermined threshold;

and store an association between the alphanumeric value and the image when the score meets or exceeds the first predetermined threshold.

9. The system of claim 8, wherein the score is based on at least one of:

confidence that the extracted alphanumeric value matches the characters in the image;

a format of the extracted alphanumeric value;

a number of digits in the extracted alphanumeric value;

a count of images in which the extracted alphanumeric value appears;

or a difference between an expected address and the extracted alphanumeric value.

10. The system of claim 8, wherein the one or more computing devices are further configured to determine whether the score is at or below a second predetermined threshold, wherein the association between the alphanumeric value and the building number is not stored if the score is at or below the second predetermined threshold.

11. The system of claim 8, wherein the one or more computing devices are further configured to request confirmation when the score is below the first predetermined threshold and above the second predetermined threshold.

12. The system of claim 8 wherein the one or more computing devices are configured to identify the portion of the image containing characters by comparing a plurality of portions of the image to image patterns of building numbers.

13. The system of claim 8 wherein the one or more computing devices are configured to extract the alphanumeric value by applying optical character recognition to the portion of the image containing characters.

14. The system of claim 8 wherein the one or more computing devices are configured to compare the alphanumeric value to the range of building numbers by comparing the alphanumeric value to a set of values that includes the building numbers of the street address.

15. The system of claim 8 wherein the one or more computing devices are configured to compare the alphanumeric value to a building number of the street address by determining whether the alphanumeric value comprises a numeric value that is equivalent to the building number of the street address.

16. The system of claim 8 wherein the one or more computing devices are further configured to transmit a proposed building number based on the alphanumeric value and the image portion in confirming whether the street level image corresponds to the street address.

17. The system of claim 8, wherein the one or more computing devices are further configured to store an association linking at least the portion of the street level image to one or more of the building numbers associated with the portion.

18. A non-transient computer-readable medium storing instructions executable by a processor to perform a method, comprising:

selecting an image associated with a street address that includes a building number;

identifying a portion of the image containing characters associated with building numbers;

extracting an alphanumeric value from the characters within the portion of the image;

comparing the alphanumeric value to the building number of the street address;

determining a score based on the comparison of the alphanumeric value to the building number of the street address;

determining whether the score meets or exceeds a first predetermined threshold; and

storing an association between the alphanumeric value and the image when the score meets or exceeds the first predetermined threshold.

19. The non-transient computer-readable medium of claim 18, wherein the score is based on at least one of:

confidence that the extracted alphanumeric value matches the characters in the image;

a format of the extracted alphanumeric value;

a number of digits in the extracted alphanumeric value;

a count of images in which the extracted alphanumeric value appears;

or a difference between an expected address and the extracted alphanumeric value.

20. The non-transient computer-readable medium of claim 18, the method further comprising determining whether the score is at or below a second predetermined threshold, wherein the association between the alphanumeric value and the building number is not stored if the score is at or below the second predetermined threshold.