Methods for mobile image capture of vehicle identification numbers in a non-document

• US 9,773,186 B2
• Filed: 03/17/2014
• Issued: 09/26/2017
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Mitek Systems Incorporated
• Original Assignee: Mitek Systems Incorporated
• Inventors: Nepomniachtchi, Grigori, Kotovich, Nikolay
• Primary Examiner(s): Johns, Andrew W
• Application Number:
  US14/217,361
• Publication Number:
  US 20140270385A1
• Time in Patent Office: 1,289 Days
• Field of Search: 382104, 382105
• CPC Class Codes: G06V 10/22 by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G06V 30/274 Syntactic or semantic context, e.g. balancing
• CPC Class Codes: G06V 30/40 Document-oriented image-based pattern recognition

Abstract

Various embodiments disclosed herein are directed to methods of capturing Vehicle Identification Numbers (VIN) from images captured by a mobile device. Capturing VIN data can be useful in several applications, for example, insurance data capture applications. There are at least two types of images supported by this technology: (1) images of documents and (2) images of non-documents.

Claims

1. A non-transitory computer readable medium containing instructions which, when executed by a computer, perform a process comprising:

receiving an image which includes a vehicle identification number (VIN);

making a color assumption with respect to the VIN;

preprocessing the image based on the color assumption;

segmenting the image to identify candidate text strings that may include the VIN;

performing an optical character recognition (OCR) using an OCR engine on the candidate text strings;

validating the candidate text strings;

if a candidate test string is validated, outputting a confirmed VIN value based on the validated candidate test string; and

if a candidate test string fails the validation, post-processing the failed candidate text string, which includes using a different OCR engine to re-recognize some or all characters in the failed candidate test string.

2. The non-transitory computer readable medium of claim 1, wherein preprocessing the image includes at least one of:

converting the image to a grayscale based on the color assumption; and

binarizing a grayscale image.

3. The non-transitory computer readable medium of claim 1, wherein validating the candidate text strings comprises performing a redundancy test.

4. The non-transitory computer readable medium of claim 3, wherein the redundancy test includes a mod11 rule test.

5. The non-transitory computer readable medium of claim 1, wherein performing an optical character recognition on the candidate text strings produces ACII text strings for each candidate text string.

6. The non-transitory computer readable medium of claim 1, wherein the color assumption may be represented by set of three weights WR, WG and WB (WR+WG+WB=1.0) to generated the color conversion formula:

BR(P)=(R(P)*WR+G(P)*WG+B(P)*WB)/3, where

P=P(X, Y)—

an arbitrary pixel on the image represented by its X and Y-coordinates,

BR(P)—

the computed brightness value of pixel P on the output grayscale image, and

R(P), G(P) and B(P)—

Red, Green and Blue color value of pixel P on the original color image.

7. A system for identifying a field in an image of a non-document, comprising:

a memory configured to store the image; and

a processor coupled with the memory, the processor configured to;

receive an image which includes a vehicle identification number (VIN);

make a color assumption with respect to the VIN;

preprocess the image based on the color assumption;

segment the image to identify candidate text strings that may include the VIN;

perform an optical character recognition (OCR) using an OCR engine on the candidate text strings;

validate the candidate text strings;

if a candidate test string is validated, output a confirmed VIN value based on the validated candidate test string; and

if a candidate test string fails the validation, post-processing the failed candidate text string, which includes using a different OCR engine to re-recognize some or all characters in the failed candidate test string.

8. The system of claim 7, wherein preprocessing the image includes at least one of:

converting the image to a grayscale based on the color assumption; and

binarizing a grayscale image.

9. The system of claim 7, wherein validating the candidate text strings comprises performing a redundancy test.

10. The system of claim 9, wherein the redundancy test includes a mod11 rule test.

11. The system of claim 7, wherein performing an optical character recognition on the candidate text strings produces ACII text strings for each candidate text string.

12. The system of claim 7, wherein the color assumption is represented by set of three weights WR, WG and WB (WR+WG +WB=1.0) to generated the color conversion formula:

BR(P)=(R(P)*WR+G(P)*WG+B(P)*WB)/3, where

P=P(X, Y)—

an arbitrary pixel on the image represented by its X and Y-coordinates,

BR(P)—

the computed brightness value of pixel P on the output grayscale image, and

R(P), G(P) and B(P)—

Red, Green and Blue color value of pixel P on the original color image.

13. A method for identifying a field in an image of a non-document, comprising:

receiving an image which includes a vehicle identification number (VIN);

making a color assumption with respect to the VIN;

preprocessing the image based on the color assumption;

segmenting the image to identify candidate text strings that may include the VIN;

performing an optical character recognition (OCR) using an OCR engine on the candidate text strings;

validating the candidate text strings;

if a candidate test string is validated, outputting a confirmed VIN value based on the validated candidate test string; and

if a candidate test string fails the validation, post-processing the failed candidate text string, which includes using a different OCR engine to re-recognize some or all characters in the failed candidate test string.

14. The method of claim 13, wherein preprocessing the image includes at least one of:

converting the image to a grayscale based on the color assumption; and

binarizing a grayscale image.

15. The method of claim 13, wherein validating the candidate text strings comprises performing a redundancy test.

16. The method of claim 15, wherein the redundancy test includes a mod11 rule test.

17. The method of claim 13, wherein performing an optical character recognition on the candidate text strings produces ACII text strings for each candidate text string.

18. The method of claim 13, wherein the color assumption may be represented by set of three weights WR, WG and WB (WR+WG+WB=1.0) to generated the color conversion formula:

BR(P)=(R(P)*WR+G(P)*WG+B(P)*WB)/3, where

P=P(X, Y)—

an arbitrary pixel on the image represented by its X and Y-coordinates,

BR(P)—

the computed brightness value of pixel P on the output grayscale image, and

R(P), G(P) and B(P)—

Red, Green and Blue color value of pixel P on the original color image.