Methods and systems for optimized parameter selection in automated license plate recognition

• US 9,292,759 B2
• Filed: 05/07/2012
• Issued: 03/22/2016
• Est. Priority Date: 05/07/2012
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Burry, Aaron Michael, Zhao, Yonghui, Kozitsky, Vladimir
• Primary Examiner(s): Koziol, Stephen R
• Assistant Examiner(s): SCHWARTZ, RAPHAEL M
• Application Number:
  US13/466,068
• Publication Number:
  US 20130294653A1
• Time in Patent Office: 1,415 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

A system and method for automatically recognizing license plate information, the method comprising receiving an image of a license plate, and generating a plurality of image processing data sets, wherein each image processing data set of the plurality of image processing data sets is associated with a score of a plurality of scores by a scoring process comprising determining one or more image processing parameters, generating the image processing data set by processing the image using the one or more image processing parameters, generating the score based on the image processing data, and associating the image processing data set with the score.

Claims

1. A method for automatically recognizing license plate information, the method comprising:

receiving an image of a license plate;

generating, using one or more processors, a first image processing data set by processing the image using a first set of image processing parameters;

generating a first computer-generated score based on the first image processing data set by;

segmenting the image into a first plurality of characters based on the first image processing data set;

generating a segment score for each character of the first plurality of characters; and

totaling the segment scores for each character of the first plurality of characters to determine the first computer-generated score;

determining that the first computer-generated score is;

greater than a first threshold score; and

less than a second threshold score;

in response to determining that the first computer-generated score is greater than the first threshold score and is less than the second threshold score, determining a second set of image processing parameters based on the first set of image processing parameters;

generating a second image processing data set by processing the image using the second set of image processing parameters;

generating a second computer-generated score based on the second image processing data set by;

segmenting the image into a second plurality of characters based on the second image processing data set;

generating a segment score for each character of the second plurality of characters; and

totaling the segment scores for each character of the second plurality of characters to determine the second computer-generated score;

determining that the second computer-generated score is greater than the second threshold score;

determining that a segment score of a character of the first plurality of characters is greater than a segment score of a corresponding character of the second plurality of characters; and

in response to determining the second computer-generated score is greater than the second threshold score and that the segment score of the character of the first plurality of characters is greater than the segment score of the corresponding character of the second plurality of characters, generating license plate data based on the character of the first plurality of characters and the second image processing data set, wherein the character of the first plurality of characters replaces the corresponding character of the second plurality of characters in the license plate data.

2. The method of claim 1, wherein generating the first image processing data set and the second image processing data set comprises performing binarization on the image.

3. The method of claim 1, wherein generating the first image processing data set and the second image processing data set comprises applying a tone reproduction curve to the image.

4. The method of claim 1, wherein generating the first image processing data set and the second image processing data set comprises applying a filter to the image.

5. The method of claim 1, wherein generating the segment score for each character of the first plurality of characters and the second plurality of characters comprises generating the segment score using a connected component analysis.

6. The method of claim 1, wherein generating the first computer-generated score and the second computer-generated score comprises generating the score based on expected license plate information.

7. The method of claim 1, wherein generating the first computer-generated score and the second computer-generated score comprises generating the score based on an expected position of license plate information.

13. The method of claim 1, further comprising generating the first set of image processing parameters based on at least one of a current time, a current traffic level, or a determined quality of the image.

14. The method of claim 1, wherein the first computer-generated score and the second computer-generated score are generated based on at least one of known properties of characters used on license plates in a determined jurisdiction, previously determined properties of previously processed license plates from the same location, or an expected number of characters on the license plate.

17. The method of claim 1, wherein the second threshold score is entered by a user prior to receiving the image of the license plate.

8. A system configured for automatically recognizing license plate information, the system comprising:

a processing system comprising one or more processors; and

a memory system comprising one or more computer-readable media, wherein the one or more computer-readable media contain instructions that, when executed by the processing system, cause the processing system to perform operations comprising;

receiving an image of a license plate;

generating a first image processing data set by processing the image using a first set of image processing parameters;

generating a first computer-generated score based on the first image processing data set by;

segmenting the image into a first plurality of characters based on the first image processing data set;

generating a segment score for each character of the first plurality of characters; and

totaling the segment scores for each character of the first plurality of characters to determine the first computer-generated score;

determining that the first computer-generated score is;

greater than a first threshold score; and

less than a second threshold score;

in response to determining that the first computer-generated score is greater than the first threshold score and is less than the second threshold score, determining a second set of image processing parameters based on the first set of image processing parameters;

generating a second image processing data set by processing the image using the second set of image processing parameters;

generating a second computer-generated score based on the second image processing data set by;

segmenting the image into a second plurality of characters based on the second image processing data set;

generating a segment score for each character of the second plurality of characters; and

totaling the segment scores for each character of the second plurality of characters to determine the second computer-generated score;

determining that the second computer-generated score is greater than the second threshold score;

determining that a segment score of a character of the first plurality of characters is greater than a segment score of a corresponding character of the second plurality of characters; and

in response to determining the second computer-generated score is greater than the second threshold score and that the segment score of the character of the first plurality of characters is greater than the segment score of the corresponding character of the second plurality of characters, generating license plate data based on the character of the first plurality of characters and the second image processing data set, wherein the character of the first plurality of characters replaces the corresponding character of the second plurality of characters in the license plate data.

9. The system of claim 8, wherein generating the first image processing data set and the second image processing data set comprises performing binarization on the image.

10. The system of claim 8, wherein generating the segment score for each character of the first plurality of characters and the second plurality of characters comprises generating the segment score using a connected component analysis.

11. The system of claim 8, wherein generating the first computer-generated score and the second computer-generated score comprises generating the score based on expected license plate information.

12. The system of claim 8, wherein generating the first computer-generated score and the second computer-generated score comprises generating the score based on an expected position of license plate information.

15. The system of claim 8, the operations further comprising generating the first set of image processing parameters based on at least one of a current time, a current traffic level, or a determined quality of the image.

16. The system of claim 8, wherein the first computer-generated score and the second computer-generated score are generated based on at least one of known properties of characters used on license plates in a determined jurisdiction, previously determined properties of previously processed license plates from the same location, or an expected number of characters on the license plate.

18. The system of claim 8, wherein the second threshold score is entered by a user prior to receiving the image of the license plate.